finding_id,source,bucket,bucket_description,finding_title,core_idea,significance,prior_expectation,support_and_caveats,research_community,agent_rank,rank_denominator,agent_novelty_grade,agent_support_grade,agent_priority_grade
bucket_1_rank_1,distributed_agents,B1,Distributed Agents catalog ranks 1-30,NAA20 perturbation couples recurrent CYP46A1 repression to induction of sterol-synthesis transcripts,"In the atlas, Naa20 CRISPR perturbation lowered Cyp46a1 in thalamic, midbrain glutamatergic, and midbrain/pons/medulla GABAergic neurons, while sterol-synthesis transcripts rose in hypothalamic and midbrain/pons/medulla contexts. The biological interpretation is that impaired NatB function may reduce neuronal cholesterol elimination through CYP46A1 and elicit compensatory sterol synthesis; the co-occurrence of Cyp46a1 repression and Fdft1 induction in 191 MB P MY GABA makes this a testable coupling rather than an inference from cross-context absence.","If confirmed at the protein and metabolite levels, this would connect NAA20-associated neurodevelopmental disease to the principal route of brain cholesterol disposal, reframing NatB pathology around a NAA20–CYP46A1–24S-hydroxycholesterol axis. That would justify an unusually broad research direction spanning NatB substrate acetylation, neuronal sterol flux, circuit consequences, disease biomarkers, and therapeutic restoration of cholesterol turnover.","Before viewing the atlas, impaired NAA20/NatB activity was expected to cause recessive neurodevelopmental disease based on human genetics and biochemical evidence (https://pubmed.ncbi.nlm.nih.gov/34230638/), while CYP46A1 was independently understood as the major route for converting brain cholesterol to 24S-hydroxycholesterol in a review of brain cholesterol turnover (https://pubmed.ncbi.nlm.nih.gov/27663182/). Those established facts did not predict that acute neuronal Naa20 perturbation would repeatedly repress Cyp46a1 or pair that response with sterol-synthesis induction.","Naa20 perturbation reduced Cyp46a1 in 151 TH Prkcd Grin2c Glut (log2FC -0.39135, FDR 1.29e-13), 155 MB Glut (-0.76576, FDR 1.03e-10), and 191 MB P MY GABA (-1.03665, FDR 2.55e-6). Sterol-pathway increases included Cyp51 in 145 MH-LH TH Glut (+0.56858, FDR 0.0272), Mvk in the same class (+1.05342, FDR 0.0339), and Fdft1 in 191 MB P MY GABA (+1.02478, FDR 1.79e-5), where Cyp46a1 was also reduced. The major alternative is an indirect stress response to NAA20 perturbation rather than a sterol-regulatory role of NatB; the experiment measured neither N-terminal acetylation, CYP46A1 protein, 24S-hydroxycholesterol, nor cholesterol flux, and the partial sterol response was not identical across neuronal classes.","NatB/N-terminal-acetylation researchers, brain-cholesterol biologists, and investigators of NAA20-related neurodevelopmental disease would have the clearest reason to revise their working models. The pivotal test is neuron-resolved Naa20 editing coupled to CYP46A1 protein and isotope-based cholesterol-to-24S-hydroxycholesterol flux, with CRISPR-resistant wild-type and catalytically inactive NAA20 resolving NatB dependence from generic editing stress. A verified causal chain would establish an unusually broad agenda spanning the neuronal N-terminal acetylome, lipid turnover, circuit dysfunction, patient metabolite biomarkers, and CYP46A1-directed rescue.",1,144,5,4,5
bucket_1_rank_2,distributed_agents,B1,Distributed Agents catalog ranks 1-30,MATR3 perturbation induces an RBP state that active FUS and TDP-43 perturbations do not reproduce,"Across cortical, thalamic, midbrain glutamatergic, and posterior GABAergic neurons, Matr3 perturbation repeatedly increased Atp9b, Ddhd1, E2f4, Kcnq2, Sugp2, Uggt2, and Zkscan3 while decreasing Rsrp1. FUS and TDP-43 perturbations generated substantial expression responses in matched classes but did not reproduce this signed MATR3 module. This observed divergence supports the interpretation that acute MATR3 reduction exposes an RNA-binding-protein liability state rather than a generic response to perturbing any ALS-linked RBP.","This relationship argues that MATR3-associated disease should not be subsumed under a single TDP-43/FUS stress model. If the module reflects direct MATR3-dependent RNA regulation, it would create a distinct mechanistic and biomarker framework for MATR3-linked ALS, frontotemporal degeneration, and distal myopathy, and would motivate comparative studies of why disease-associated RBPs converge in some settings yet diverge sharply in mature neurons.","MATR3 was already known as a nuclear RNA-binding protein associated with neuromuscular disease and selective neuronal vulnerability (https://doi.org/10.1172/jci.insight.143948), with additional evidence connecting its perturbation to neurodegenerative RNA dysregulation (https://doi.org/10.1038/s41467-020-18949-w). Recent work also provided precedent for convergence among ALS-associated RNA-binding proteins on a REST–UNC13A mechanism (https://doi.org/10.1038/s44318-025-00506-0), so a shared acute response would have been reasonable before the atlas; the recurrent, oppositely signed MATR3 module was not.","In 005 L4-5 IT CTX Glut, Matr3 perturbation increased Atp9b (+1.06061, FDR 4.14e-12), Ddhd1 (+1.59396, FDR 4.23e-14), and E2f4 (+1.29225, FDR 4.72e-8) and decreased Rsrp1 (-0.35285, FDR 1.96e-4). The same directions were stronger in 151 TH Prkcd Grin2c Glut: Atp9b +1.48175 (FDR 3.53e-38), Ddhd1 +1.76597 (FDR 2.64e-49), E2f4 +1.73480 (FDR 1.46e-35), and Rsrp1 -0.69262 (FDR 1.26e-29). In 155 MB Glut, Ddhd1 rose +1.93699 (FDR 2.84e-28) and Rsrp1 fell -0.51681 (FDR 1.09e-14); in 191 MB P MY GABA, Ddhd1 (+1.89191, FDR 1.43e-11), E2f4 (+2.23139, FDR 1.42e-11), and Atp9b (+0.94914, FDR 0.00455) remained significant, although Rsrp1 did not (FDR 0.377). By comparison, Fus was strongly reduced in 151 (-1.40691, FDR 1.04e-45) yet Atp9b and Ddhd1 had FDR 1.0 and Rsrp1 moved oppositely upward (+0.42834, FDR 4.84e-29); Tardbp was likewise reduced (-1.41422, FDR 1.60e-39) while Rsrp1 increased (+0.15504, FDR 3.24e-5). The chief limitation is that guide efficiency and acute perturbation biology can generate apparent divergence; gene-level snRNA-seq does not measure the splicing, RNA export, binding occupancy, or aggregation mechanisms expected of these proteins.","ALS/FTD and neuromuscular-disease researchers—especially those studying MATR3 and comparative RBP biology—would be most likely to change direction. Matched-efficiency MATR3, FUS, and TARDBP editing in the same neuron types, followed by long-read RNA sequencing, MATR3 eCLIP, and CRISPR-resistant MATR3 rescue, would determine whether the signed module is a direct MATR3 regulon or a product of unequal perturbation or generic stress. Direct target recovery would justify MATR3-specific vulnerability maps, molecular biomarkers, and RNA-targeted interventions instead of a universal ALS-RBP model.",2,144,5,4,4
bucket_1_rank_3,distributed_agents,B1,Distributed Agents catalog ranks 1-30,PPP2R1A perturbation reverses sterol-transcription polarity across neuronal classes,Ppp2r1a perturbation induced multiple cholesterol-biosynthesis genes in midbrain and hypothalamic glutamatergic neurons but reduced Srebf2 and Hmgcr in thalamic and cortical glutamatergic neurons. The observation therefore supports a testable interpretation in which the PP2A A-alpha scaffold gates qualitatively different sterol-homeostatic states across mature neuronal classes rather than producing one uniform cholesterol response.,"A validated polarity switch would make PP2A holoenzyme context an important determinant of neuronal lipid regulation, connecting phosphatase composition to opposing SREBP-pathway outputs. This could launch mechanistic studies of which PP2A complexes, phosphorylation events, and neuronal states determine sterol synthesis, and could alter how PPP2R1A-associated neurodevelopmental phenotypes or PP2A-directed interventions are interpreted.","A curated cholesterol-biosynthesis pathway (https://reactome.org/content/detail/R-HSA-191273) places SREBP2/HMGCR control and downstream enzymes such as MSMO1, DHCR24, and SC5D within a coordinated sterol program. Before the atlas, a perturbation affecting that pathway would more naturally have been expected to shift the module in a common direction; prior knowledge did not predict opposite PPP2R1A-associated RNA outputs among these neuronal classes.","After Ppp2r1a perturbation, 155 MB Glut neurons increased Msmo1 (+1.01390, FDR 2.95e-9), Insig1 (+0.93877, FDR 4.72e-7), Dhcr24 (+0.81862, FDR 7.79e-7), and Sc5d (+0.75365, FDR 1.10e-6). The 110 CNU-HYa HY MM Glut class showed the same direction for Insig1 (+0.93884, FDR 1.69e-5), Sc5d (+0.81201, FDR 1.42e-4), Dhcr24 (+0.76304, FDR 2.67e-4), and Msmo1 (+0.82738, FDR 6.80e-4). Conversely, 151 TH Prkcd Grin2c Glut reduced Srebf2 (-0.53092, FDR 9.47e-8) and Hmgcr (-0.32009, FDR 0.0291), while 005 L4-5 IT CTX Glut reduced Hmgcr (-0.59273, FDR 3.17e-5) and Srebf2 (-0.27908, FDR 0.0414). Because different genes mark the two arms, this is a pathway-level polarity interpretation rather than a literal sign reversal of every transcript; PP2A holoenzyme composition, phosphoproteins, SREBP2 cleavage, and sterol abundance were not measured, and unequal editing or baseline state remains an alternative.","PP2A signaling researchers, neuronal lipid biologists, and investigators of PPP2R1A-related developmental disorders are the natural audience. Their critical experiment is matched Ppp2r1a editing in purified representatives of the opposing neuronal classes, with phosphoproteomics, nuclear SREBP2 quantification, isotope-resolved sterol synthesis, and wild-type rescue. Opposite biochemical flux at comparable engagement would establish a phosphatase-dependent switch and motivate systematic mapping of PP2A holoenzymes to lipid-control nodes and neuronal physiology.",3,144,4,4,4
bucket_1_rank_4,distributed_agents,B1,Distributed Agents catalog ranks 1-30,AP2M1 and AP2S1 perturbations converge on TFRC induction and sterol repression while AP2S1 adds cortical activity-gene loss,"Perturbing either of two physical AP-2 subunits induced Tfrc and reduced sterol-regulatory transcripts in matched thalamic and midbrain neuronal classes; Ap2s1 perturbation additionally repressed an Arc/Egr3/Bdnf/Vgf branch in cortical neurons. Together, these observations suggest that neuronal AP-2 dysfunction may couple impaired cargo endocytosis to an iron-uptake compensation signal and lipid regulation, with a further AP2S1-associated activity-output branch.","The convergence links a core endocytic complex to coordinated neuronal iron-receptor, sterol, and excitability-associated RNA outputs rather than treating AP-2 disease solely as defective vesicle trafficking. If validated, it would create a tractable iron–lipid–activity framework for AP2M1 developmental and epileptic encephalopathy and related AP-2 disorders, with distinct opportunities for cargo-level and neuronal-class interventions.","AP2M1 and AP2S1 were already established as physical subunits of the AP-2 clathrin adaptor (https://rest.uniprot.org/uniprotkb/P84091), and AP-2 depletion was known to impair transferrin-receptor uptake (https://pubmed.ncbi.nlm.nih.gov/12952941/). AP2M1 dysfunction is also linked to developmental and epileptic encephalopathy (https://pubmed.ncbi.nlm.nih.gov/31104773/). Those findings made defective endocytosis expected, but did not predict shared neuronal Tfrc induction with sterol repression or the AP2S1-associated cortical activity-gene branch.","In 151 TH Prkcd Grin2c Glut, Ap2m1 perturbation induced Tfrc (+0.83140, FDR 3.13e-6), and Ap2s1 perturbation induced it even more strongly (+0.98358, FDR 3.06e-24); both target transcripts were reduced in this class (Ap2m1 -0.57819, FDR 2.41e-8; Ap2s1 -0.53643, FDR 6.33e-5). In 155 MB Glut, Ap2m1 perturbation increased Tfrc (+0.96399, FDR 1.07e-5) while reducing Srebf2 (-0.78418, FDR 3.60e-5) and Dhcr24 (-0.83428, FDR 0.00323); Ap2s1 perturbation showed the same pattern for Tfrc (+0.85038, FDR 1.00e-7), Srebf2 (-0.49178, FDR 0.00229), and Dhcr24 (-0.55046, FDR 0.0146). The additional 005 L4-5 IT CTX Glut Ap2s1 response included Egr3 -1.33292 (FDR 1.35e-13), Arc -0.89254 (FDR 2.56e-7), Bdnf -1.10465 (FDR 0.0274), and Vgf -0.63317 (FDR 2.52e-4). Convergence of two subunits argues against a single-guide artifact, but the assay did not measure coated pits, surface TFRC, iron uptake, cholesterol, or firing; generic membrane stress and unequal perturbation strength remain alternatives.","Endocytosis and membrane-trafficking researchers, neuronal iron and cholesterol biologists, and AP2M1 encephalopathy investigators have the strongest stake in this relationship. Comparing AP2M1- and AP2S1-edited neurons by live transferrin internalization, surface-TFRC imaging, labile-iron measurements, SREBP2 processing, lipidomics, and subunit-specific rescue would distinguish compensation for AP-2 cargo failure from a generic transcriptional stress state. The resulting cargo-to-state map could organize work on neuronal lipid balance, excitability, and genotype-specific therapeutic vulnerabilities.",4,144,4,4,4
bucket_1_rank_7,distributed_agents,B1,Distributed Agents catalog ranks 1-30,PGK1 and GAPDH perturbations converge on OGT and nuclear-speckle transcript loss,"Independent Pgk1 and Gapdh perturbations reduced Ogt together with Pnisr/Rsrp1-associated RNA-processing transcripts in thalamic glutamatergic neurons, and Tpi1 perturbation also reduced Ogt in hypothalamic glutamatergic neurons. These observations motivate, but do not establish, a mechanism linking adjacent lower-glycolysis steps to O-GlcNAc regulation and nuclear-speckle or RNA-processing maintenance; the response is not demonstrated for every glycolytic enzyme.","This convergence offers a specific route by which glycolytic insufficiency could remodel neuronal RNA processing rather than merely lowering cellular energy. If flux, O-GlcNAc signaling, and splice regulation prove causally linked, the finding would support a substantial metabolism-to-RNA-biology program relevant to neuronal stress and metabolic neurological disease.","GAPDH and PGK1 were known as adjacent enzymes in cytosolic glycolysis (https://reactome.org/content/detail/R-HSA-70171), and TPI1 as the enzyme interconverting DHAP and G3P in the glycolytic pathway (https://reactome.org/content/detail/R-HSA-70454). PNISR, RSRP1, SRRM2, and related readouts had curated nuclear-speckle or spliceosome annotations (https://www.uniprot.org/). Before the atlas, these facts did not predict that distinct lower-glycolysis perturbations would converge on Ogt and a compact RNA-processing transcript set in neurons.","In 151 TH Prkcd Grin2c Glut, Pgk1 perturbation reduced its own transcript (-1.44092, FDR 4.43e-32), Pnisr (-0.22694, FDR 7.00e-11), and Ogt (-0.14565, FDR 4.998e-4). Gapdh perturbation in the same class independently reduced Pnisr (-0.25522, FDR 2.19e-5), Rsrp1 (-0.19735, FDR 1.49e-4), and Ogt (-0.17676, FDR 0.0220). Tpi1 perturbation reduced Ogt in 145 MH-LH TH Glut (-0.38744, FDR 0.0121), although its secondary 151 RNA-processing changes were only nominal. GAPDH itself was absent from the measured features, metabolic flux and UDP-GlcNAc/O-GlcNAc levels were not assayed, and the modest effect sizes leave shared energy stress or altered cell state as important alternatives to a specific glycolysis–OGT mechanism.","Neuro-metabolism, O-GlcNAc, and RNA-splicing communities—particularly investigators of metabolic control over nuclear organization—would pursue this connection. Isotope tracing, ATP and UDP-GlcNAc measurements, quantitative O-GlcNAc proteomics, live speckle imaging, and long-read RNA sequencing after matched PGK1 and GAPDH perturbations would locate the defect; restoration by a defined metabolite or OGT rescue would separate causal signaling from nonspecific stress. The larger opportunity is a map of metabolic nodes that govern neuronal RNA processing and create disease-selective liabilities.",7,144,4,4,4
bucket_1_rank_8,distributed_agents,B1,Distributed Agents catalog ranks 1-30,"STXBP1 perturbation produces distinct cortical, midbrain, and Pvalb transcriptional outputs","Stxbp1 perturbation elicited a cortical Nrsn1/Pcsk1n-high and Grin2b-low response, a midbrain Cntnap1-low and Mt1/Fos-high response, and Slc6a1 repression in Pvalb GABA neurons. The biological interpretation is that chronic Munc18-1 insufficiency may be translated into different adaptive or stress states across neuronal classes rather than one transcriptomic consequence of presynaptic failure.","STXBP1 disorders are often framed around a common vesicle-release defect; these distinct positive RNA outputs nominate neuronal-class-specific routes from that shared lesion to circuit pathology. Validation could redirect disease studies toward differential cortical, midbrain, and inhibitory-neuron compensation and create class-specific biomarkers or therapeutic endpoints for STXBP1 encephalopathy.","STXBP1/Munc18-1 was established as a syntaxin-binding Sec1/Munc18 protein (https://rest.uniprot.org/uniprotkb/P61764) required for neurotransmitter release (https://doi.org/10.1126/science.287.5454.864), and SNARE-complex biology (https://pubmed.ncbi.nlm.nih.gov/19164740/) predicted severe presynaptic dysfunction after its loss. That prior biology predicted a shared release defect, but not the distinct cortical neurosecretory, midbrain stress/adhesion, and Pvalb transporter RNA outputs observed here.","In 005 L4-5 IT CTX Glut, Stxbp1 perturbation reduced Stxbp1 (-0.520, FDR 9.66e-11) and Grin2b (-0.231, FDR 0.00582) while increasing Nrsn1 (+0.724, FDR 3.98e-4) and Pcsk1n (+0.448, FDR 5.87e-4). In 155 MB Glut, Stxbp1 was reduced (-0.608, FDR 2.76e-9), Cntnap1 decreased (-0.526, FDR 3.71e-5), and Mt1 (+1.107, FDR 9.47e-4) and Fos (+1.390, FDR 0.0464) increased. In 052 Pvalb Gaba, Stxbp1 perturbation lowered Slc6a1 (-0.412, FDR 1.27e-8); direct Slc6a1 perturbation strongly reduced its own transcript (-1.018, FDR 1.39e-29) without establishing that the broader STXBP1 mechanism runs through SLC6A1. The atlas measured steady-state nuclear RNA rather than release, firing, survival, or network activity, so different activity histories, effective power, and perturbation strength could explain part of the apparent class divergence.","STXBP1 encephalopathy researchers, synaptic-release biologists, and circuit physiologists have the clearest incentive to follow this finding. In genetically identified cortical, midbrain, and Pvalb neurons, optical release measurements, patch clamp, time-resolved single-cell RNA, and STXBP1 restoration should determine whether the RNA branches persist at matched degrees of release failure. A true class-dependent adaptation would provide a framework linking distinct molecular responses to seizure, movement, and cognitive phenotypes and to tailored rescue strategies.",8,144,4,4,4
bucket_1_rank_10,distributed_agents,B1,Distributed Agents catalog ranks 1-30,Acute TBR1 perturbation recurrently raises CPNE6 while splitting mature cortical projection-neuron outputs,"Acute Tbr1 perturbation increased Cpne6 in several cortical projection-neuron classes, coupled to Dkk3/Epb41 decreases in L5/6 intratelencephalic neurons but to Cacna1g/C1qtnf4 increases in NP-CT-L6b-OB neurons. A plausible interpretation is that TBR1 retains a mature synaptic and membrane-associated regulon whose output depends on projection-neuron identity and is separable from its canonical neonatal fate-specification role.","A direct adult TBR1 regulon would extend TBR1 biology beyond developmental layer specification and provide a route from a neurodevelopmental transcription factor to ongoing projection-neuron physiology. This could launch focused studies of adult TBR1 chromatin occupancy, synaptic maintenance, and whether disease alleles perturb mature functions independently of early cortical wiring.","Established work positioned TBR1 as a regulator of cortical projection-neuron and layer-6 identity (https://pubmed.ncbi.nlm.nih.gov/11239428/), showed continuing dosage-sensitive roles in postnatal cortical neurons (https://pubmed.ncbi.nlm.nih.gov/30318412/), and linked TBR1 mutations to autism (https://pubmed.ncbi.nlm.nih.gov/25232744/). Before viewing the atlas, one might therefore have expected acute postnatal perturbation either to reuse familiar developmental targets or to have little mature-neuron output; a recurrent Cpne6-centered, class-split response was not the standard prediction.","In 001 L5-6 IT Glut, Tbr1 perturbation decreased Dkk3 (-1.17698, FDR 0.00229) and Epb41 (-1.54013, FDR 0.00724) while increasing Cpne6 (+0.91271, FDR 0.00748). In 005 L4-5 IT CTX Glut, Cpne6 again increased (+0.60600, FDR 0.0237) and Dkk3 decreased (-0.71992, FDR 0.0360). By contrast, the 027 NP-CT-L6b-OB Glut response was all upward among the reported rows, including Cacna1g +1.43196 (FDR 1.12e-5), Cpne6 +1.24500 (FDR 5.98e-5), and C1qtnf4 +0.47990 (FDR 0.0249). These are small expression programs from unequally sampled classes, and the assay did not measure TBR1 occupancy or direct transcription; residual lineage state, editing differences, or an indirect activity response could explain the split, while absence of neonatal targets is only a developmental comparison rather than a matched negative perturbation.","Cortical-development, projection-neuron, TBR1-syndrome, and activity-dependent-transcription communities would be most interested. Adult-neuron TBR1 CUT&RUN or CUT&Tag paired with nascent RNA after acute editing in the two responsive projection classes—and then CRISPR-resistant rescue—can distinguish direct occupancy at Cpne6, Dkk3/Epb41, or Cacna1g/C1qtnf4 from secondary state change. That result would let the field separate developmental from maintenance functions of TBR1 and test how patient variants alter adult cortical circuits.",10,144,4,4,4
bucket_1_rank_11,distributed_agents,B1,Distributed Agents catalog ranks 1-30,"Zbtb11 perturbation recurrently lowers Rbm28, Stx16, Acsf3, and Ndufaf1 across subcortical neurons","A recurrent downregulated core followed Zbtb11 perturbation in 151 TH Prkcd Grin2c Glut, 155 MB Glut, and 191 MB P MY GABA neurons, spanning Acsf3 and Ndufaf1, the nucleolar factor Rbm28, the trafficking factor Stx16, and the GPI-biosynthesis gene Pigq. This observation raises the biological interpretation that functional ZBTB11 disruption couples nuclear control of mitochondrial capacity to nucleolar, glycan, and endomembrane maintenance rather than affecting mitochondrial transcripts alone.","If this coupling is real, it would give ZBTB11-associated intellectual disability a tractable multi-compartment neuronal mechanism and justify a substantial program linking mitochondrial output, ribosome biogenesis, membrane trafficking, and glycan biology. The atlas adds a recurrent neuronal RNA state and specific readouts for testing how ACSF3-related biochemistry fits into the broader syndrome.","ZBTB11 was already expected to support nuclear mitochondrial gene expression through NRF-2/GABP, and biallelic variants were linked to intellectual disability and altered ACSF3 biochemistry (https://pubmed.ncbi.nlm.nih.gov/35104841/). Its broad tissue expression (https://www.proteinatlas.org/ENSG00000066422-ZBTB11) made a narrowly mitochondrial effect plausible but did not predict the recurrent Rbm28-Stx16-Pigq branch.","Zbtb11 perturbation yielded 48 FDR-significant genes in 151 TH Prkcd Grin2c Glut, 17 in 191 MB P MY GABA, and 16 in 155 MB Glut. In 151 TH, Rbm28 decreased by log2FC -1.460 (FDR 3.57e-16), Stx16 by -1.320 (FDR 2.29e-14), Pigq by -1.260 (FDR 2.03e-13), and Acsf3 by -1.146 (FDR 7.68e-09); Stx16 also decreased in 155 MB Glut by -1.731 (FDR 1.50e-13), while Acsf3 and Ndufaf1 decreased in 191 MB P MY GABA by -1.897 (FDR 1.33e-06) and -2.459 (FDR 8.18e-04), respectively. The response burden covaries with perturbed-cell count, the relevant GABP partners were not perturbed, and RNA abundance alone cannot distinguish direct ZBTB11 targets from secondary mitochondrial or nucleolar stress.","Neurodevelopmental-disease, mitochondrial-gene-regulation, and nucleolar-stress researchers have the strongest reason to revise their models. Acute ZBTB11 protein depletion and rescue in matched neuronal cultures, measured alongside ZBTB11/GABP occupancy, nascent transcription, pre-rRNA synthesis, and respiration, would reveal whether the multi-compartment state begins with direct coordinated control or secondary stress. Early occupancy-linked effects would anchor a ZBTB11 disease program connecting genotype, ACSF3 metabolism, ribosome production, membrane trafficking, and circuit physiology.",11,144,4,4,4
bucket_1_rank_14,distributed_agents,B1,Distributed Agents catalog ranks 1-30,"Arid1a, Smarcb1, Smarcc2, and Smarce1 perturbations converge on Ptpn3-Reep2-Prkcd repression with Clstn2-Nexn induction","Within 151 TH Prkcd Grin2c Glut neurons, perturbing four BAF subunits repeatedly lowered Ptpn3, Reep2, Prkcd, Begain, and Lrrc7 while raising Clstn2 or Nexn; Arid1b and the SMARCA ATPase perturbations did not reproduce the corrected multi-gene pattern. One interpretation is dependence on a particular mature-neuronal cBAF/core configuration, although unequal functional disruption could also create the apparent subunit boundary.","If the convergence reflects mature cBAF architecture, it would extend developmental BAF-switch models into maintenance of an adult neuronal identity and excitability program. That would create substantial follow-up directions for Coffin-Siris and related chromatin-remodeling disorders, including which subunits are mechanistically interchangeable after neuronal maturation and which outputs are reversible.","Canonical cBAF-complex assembly (https://reactome.org/content/detail/R-HSA-9933937) places ARID1A, SMARCB1, SMARCC2, and SMARCE1 in related complexes, with ARID1A and ARID1B as alternative subunits. Neural BAF complexes were already known to undergo developmental subunit switching (https://pubmed.ncbi.nlm.nih.gov/17640523/) and to control dendritic programs in postmitotic neurons (https://pubmed.ncbi.nlm.nih.gov/17920018/). Some convergence was therefore expected, but the exact signed output and incomplete paralog phenocopy were not.","Under Arid1a perturbation in 151 TH Prkcd Grin2c Glut, Begain fell by log2FC -0.455 (FDR 4.82e-06), Kcnq2 by -0.219 (FDR 1.51e-04), Ptpn3 by -0.187 (FDR 0.00302), and Lrrc7 by -0.128 (FDR 0.00463), while Clstn2 rose by +0.385 (FDR 4.82e-06). Smarcb1 perturbation lowered Ptpn3 by -0.952 (FDR 1.30e-31), Reep2 by -0.610 (FDR 5.54e-26), and Prkcd by -0.537 (FDR 2.45e-12), while increasing Nexn by +0.652 (FDR 4.94e-09). Smarcc2 produced the same signs for Ptpn3 (-0.982, FDR 6.88e-24), Reep2 (-0.681, FDR 2.02e-20), Prkcd (-0.575, FDR 1.55e-10), and Nexn (+0.644, FDR 2.71e-06); Smarce1 likewise lowered Ptpn3 (-0.340, FDR 1.23e-07), Reep2 (-0.247, FDR 1.23e-07), and Prkcd (-0.357, FDR 6.58e-07). Arid1b and Smarca2 transcripts were significantly reduced yet the same marker panel was not FDR-supported, but protein loss and BAF occupancy were unmeasured, so this boundary cannot by itself prove true subunit dispensability. This matched within-context convergence is informative, but observing it only in high-yield 151 does not establish cell-type specificity.","Neuronal chromatin-remodeling, BAF-complex, and Coffin-Siris syndrome researchers would be most likely to change their working models. Matched acute degradation of each endogenous subunit in one mature thalamic model, followed by rescue, complex proteomics, chromatin accessibility, and occupancy at the Ptpn3/Reep2/Prkcd and Clstn2/Nexn loci, would separate complex-architecture dependence from unequal editing. A defined cBAF output module would underpin studies of mature-neuron complex composition, circuit excitability, disease-variant stratification, and postnatal reversibility.",14,144,4,4,4
bucket_1_rank_17,distributed_agents,B1,Distributed Agents catalog ranks 1-30,Ranbp2 perturbation produces Srrm2-down/Rbm15-up cortical and Grm1-Epha4-Itpr1-up cerebellar responses,Ranbp2 perturbation produced a mixed RNA and nuclear-protein abundance response in 022 L5 ET CTX Glut neurons but a distinct Grm1-Epha4-Itpr1 calcium and synaptic induction in 308 CB GABA neurons. The cell-state interpretation is that the same RANBP2 nuclear-pore/SUMO lesion may be translated into different downstream liabilities by cortical projection neurons and cerebellar inhibitory neurons.,"A validated bifurcation would change RANBP2 models from a uniform nuclear-transport defect to a neuron-state-dependent mechanism with separable RNA and calcium consequences. It would create substantial opportunities to connect nuclear-pore SUMOylation to projection-neuron maintenance, cerebellar signaling, and the selective neurological manifestations of RANBP2 dysfunction.","RANBP2 was already known as a nuclear-envelope component of the RanBP2–RanGAP1–SUMO1–Ubc9 complex (https://reactome.org/content/detail/R-HSA-3108232) with SUMO E3-ligase activity (https://pubmed.ncbi.nlm.nih.gov/22464730/), so nuclear and RNA-related effects were plausible. A cerebellar calcium and synaptic induction was not an obvious generic consequence of that established function.","In 022 L5 ET CTX Glut, Ranbp2 itself decreased by log2FC -0.536 (FDR 0.00653), Gpr162 decreased by -0.487 (FDR 0.00778), Rbm15 increased by +0.751 (FDR 0.0194), and Srrm2 decreased by -0.373 (FDR 0.0350). In 308 CB GABA, Grm1 increased by +0.829 (FDR 0.00872), Epha4 by +1.084 (FDR 0.00872), and Itpr1 by +1.412 (FDR 0.0261). The two positive responses occur outside the highest-yield contexts, but the atlas measures RNA rather than SUMOylation, transport, or calcium physiology; unequal protein disruption or baseline state could therefore create two apparent branches without one common causal RANBP2 mechanism.","Nuclear-transport, SUMOylation, cortical-projection-neuron, and cerebellar-physiology researchers would be most affected. Matched acute RANBP2 protein depletion in L5 ET-like and cerebellar GABA neurons, with quantitative SUMO proteomics and transport assays in both and calcium imaging plus electrophysiology in the latter, would test for equal proximal disruption followed by divergent distal phenotypes. Such state-dependent decoding would motivate cell-class-specific models of RANBP2 circuit dysfunction and selective rescue.",17,144,4,4,4
bucket_1_rank_18,distributed_agents,B1,Distributed Agents catalog ranks 1-30,Rb1 and E2f4 perturbations release distinct replication and chromosome-repair arms in thalamic neurons,"In 151 TH Prkcd Grin2c Glut neurons, Rb1 perturbation induced a Dnmt1-Tipin-Prim1-Lig1 DNA-replication arm, whereas E2f4 perturbation induced an Aaas-Rad18-Tacc3-Blm-Ddx11-Polq chromosome and repair arm. The observed separation supports a model in which postmitotic neurons retain distinct RB1-E2F and DREAM-E2F4 brakes, making ectopic cell-cycle transcription a set of replication-licensing and chromosome-maintenance routes rather than one undifferentiated re-entry program.","Resolving these arms could materially sharpen models of abortive neuronal cell-cycle re-entry in degeneration and injury. It would support substantial work on which branch causes DNA damage, apoptosis, or survival and whether selective suppression of one arm can protect mature neurons without broadly blocking cell-cycle machinery elsewhere.","Acute Rb inactivation was already known to trigger ectopic cell-cycle proteins and neuronal loss (https://pubmed.ncbi.nlm.nih.gov/23077065/), while RB–E2F and DREAM–E2F4 complexes were known to repress related but separable cell-cycle programs (https://pubmed.ncbi.nlm.nih.gov/35835684/) and E2F4 was established as a p130–DREAM repressor component (https://pubmed.ncbi.nlm.nih.gov/17531812/). Thus derepression was expected, while the clean replication-versus-repair partition in this neuronal state was not.","After Rb1 perturbation in 151 TH Prkcd Grin2c Glut, Dnmt1 increased by log2FC +1.514 (FDR 9.33e-10), Tipin by +1.518 (FDR 2.93e-06), Prim1 by +2.724 (FDR 5.87e-05), and Lig1 by +1.942 (FDR 1.63e-04); Dnmt1 also rose in 155 MB Glut by +1.158 (FDR 7.69e-04). After E2f4 perturbation in 151 TH, Aaas increased by +1.331 (FDR 3.65e-14), Rad18 by +1.635 (FDR 5.31e-06), and Polq by +2.120 (FDR 0.00763). E2f1 and E2f5 did not reproduce these active states, but their negative comparisons do not establish functional equivalence of editing. Rb1 self-RNA was not FDR-reduced, and the assay does not measure RB1 protein, DREAM occupancy, DNA synthesis, DNA damage, apoptosis, or actual cell-cycle progression. The within-151 contrast is valid, but its high effective power precludes an inference that these arms are thalamus-specific.","Neuronal cell-cycle, DNA-damage, and neurodegeneration communities would be most motivated. Matched acute RB1 and E2F4 protein depletion in mature neurons should combine nascent RNA and RB/E2F4/p130 occupancy with EdU, replication-fork markers, DNA-damage imaging, and survival. Distinguishing transcriptional derepression from abortive S phase—and assigning different physiological outcomes to the two arms—would support branch-selective intervention across neurodegenerative and acute neuronal-injury models.",18,144,3,4,4
bucket_1_rank_19,distributed_agents,B1,Distributed Agents catalog ranks 1-30,Tardbp and Fus perturbations create opposite Snrnp70 responses within separable recurrent RNA-processing states,"Across 151 TH Prkcd Grin2c Glut, 155 MB Glut, and 191 MB P MY GABA neurons, Tardbp perturbation recurrently induced Ppp6c, Fam135a, Gabpa, and Usp15 while lowering Snrnp70; Fus perturbation instead induced Rsrp1 and Atxn2l and raised Snrnp70. These opposing, recurrent observations support distinct acute TDP-43 and FUS RNA-processing abundance states rather than one generic ALS-gene stress response.","The signed divergence offers a way to separate TDP-43 and FUS mechanisms in matched neurons and could yield state-specific biomarkers or modifiers for two major RNA-binding-protein disease axes. Because the signal recurs in several neuronal contexts, confirmation could sustain substantial programs on how shared RNA-processing machinery generates distinct downstream toxicity.","TDP-43 was already established as a repressor of cryptic exons (https://pubmed.ncbi.nlm.nih.gov/26250685/) with cell-type-dependent RNA targets (https://pubmed.ncbi.nlm.nih.gov/28153034/), and FUS was known to bind brain RNAs and regulate alternative splicing (https://doi.org/10.1038/srep00603). These priors predicted RNA-processing defects but not the recurrent opposite Snrnp70 direction or the exact Tardbp-versus-Fus abundance programs.","In 151 TH Prkcd Grin2c Glut, Tardbp decreased by log2FC -1.414 (FDR 1.60e-39), while Ppp6c, Fam135a, and Gabpa increased by +1.985, +2.132, and +2.248 (FDR 2.45e-41, 4.16e-35, and 1.74e-38) and Snrnp70 decreased by -0.587 (FDR 5.10e-20). The signs recurred in 155 MB Glut for Ppp6c (+2.264, FDR 5.11e-30) and Snrnp70 (-0.667, FDR 2.80e-14), and in 191 MB P MY GABA for Ppp6c (+2.463, FDR 3.90e-17) and Snrnp70 (-0.589, FDR 1.53e-06). Fus perturbation in 151 TH lowered Fus by -1.407 (FDR 1.04e-45) but increased Rsrp1 by +0.428 (FDR 4.84e-29), Atxn2l by +0.411 (FDR 3.84e-14), and Snrnp70 by +0.261 (FDR 5.74e-05); Rsrp1 and Snrnp70 were also both positive in 155 MB and 191 MB. The 3-prime abundance assay cannot observe cryptic-exon inclusion, Stmn2 was not FDR-significant in the 151 Tardbp comparison, Unc13a abundance increased rather than recapitulating human cryptic-exon-associated depletion, and weak negative ALS comparators do not exclude their disease mechanisms.","ALS/FTD, RNA-splicing, and neuronal RNP communities would be most engaged. Isoform-resolved long-read or junction-level single-nucleus sequencing after matched acute TDP-43 and FUS depletion, combined with eCLIP and RNA-binding-mutant rescue, can separate direct isoform regulation from secondary abundance compensation at Snrnp70 and companion genes. Mechanistic separation would enable comparative biomarkers, modifier screens, and therapy testing tailored to TDP-43 versus FUS states.",19,144,3,4,4
bucket_1_rank_21,distributed_agents,B1,Distributed Agents catalog ranks 1-30,Proteasome-core and assembly perturbations couple proteasome-gene induction to loss of neuronal ion-pump and glutamate-receptor transcripts,"Perturbing Psmb4, Psmc1, Psmc5, or Pomp repeatedly increased proteasome or proteostasis transcripts while lowering Atp1a3, Atp1b1, or Gria2 in responsive midbrain glutamatergic, Pvalb, and hypothalamic GABA neurons. The interpretation is not that proteasome function was restored, but that compensatory transcription coexists with erosion of a mature-neuronal expression state.","If the two arms are causally linked, proteasome stress in neurons is not adequately described by the familiar recovery response: the same cells may restore proteasome-component RNA while losing molecular support for membrane potential and glutamatergic signaling. That would justify a substantial program to define when proteasome compensation preserves neurons and when it marks entry into a physiologically compromised state relevant to neurodegeneration.","Proteasome impairment was expected to engage recovery pathways because POMP was established as a proteasome-assembly factor (https://pubmed.ncbi.nlm.nih.gov/17948026/), neuronal loss of PSMC1 was known to cause proteasome failure and neurodegeneration (https://www.jneurosci.org/content/28/33/8189), and NFE2L1 had been shown to mediate proteasome recovery after inhibition (https://pubmed.ncbi.nlm.nih.gov/20385086/). Those sources do not predict the accompanying loss of ion-pump and glutamate-receptor RNA.","In 155 MB Glut, Pomp perturbation lowered Atp1a3 (log2FC -1.075, FDR 2.98e-11) and Gria2 (-0.997, FDR 1.16e-6) while increasing Ddit3 (+1.674, FDR 3.04e-5) and Psmc5 (+0.530, FDR 0.00268). In the same class, Psmc1 perturbation lowered Atp1a3 (-1.432, FDR 1.62e-23) and Atp1b1 (-1.556, FDR 1.98e-25), increased Ddit3 (+2.122, FDR 1.32e-9) and Psmd4 (+0.921, FDR 1.01e-7); Psmc5 perturbation likewise lowered Atp1a3 (-1.247, FDR 8.07e-13) and Atp1b1 (-1.094, FDR 1.78e-8) while increasing Psma5 (+0.891, FDR 0.00244). In 052 Pvalb Gaba, Psmb4 perturbation lowered Atp1a3 (-1.001, FDR 3.49e-4) and Gria2 (-0.886, FDR 1.20e-5) while increasing Psmc6 (+0.766, FDR 2.39e-6), Psmd4 (+0.960, FDR 5.34e-5), and Psme4 (+0.713, FDR 2.28e-4); the same directions recurred in 066 CNU-HYa HY GABA. The principal limitation is that RNA abundance does not measure proteasome activity, recovery, neuronal firing, or survival; shared integrated stress or a general state shift could generate both arms, and functional editing was not independently quantified.","Neuronal proteostasis, proteasome-recovery, and neurodegeneration researchers need to distinguish compensatory RNA induction from physiological rescue. Matched editing of several proteasome nodes in defined neurons, with proteasome-activity reporters, longitudinal survival, and patch-clamp measurement of Na+/K+-pump and glutamatergic function, would reveal whether recovery-gene induction can be uncoupled from decline. That distinction could guide a subunit-by-neuron vulnerability program and interventions that preserve excitability during proteasome stress.",21,144,3,4,4
bucket_1_rank_22,distributed_agents,B1,Distributed Agents catalog ranks 1-30,"ANK3 perturbation lowers calcium-handling, receptor, and active-zone transcripts without a matched Cntn1, Cntnap1, or Sptan1 response","ANK3-targeted thalamic Prkcd/Grin2c glutamatergic neurons lost a broad set of calcium-handling, synaptic, and receptor transcripts, whereas targeting the nodal or cytoskeletal partners Cntn1, Cntnap1, and Sptan1 did not reproduce that RNA state in the same context. This matched observation supports, without proving, a model in which ankyrin-G maintains a transcriptionally coupled excitability program beyond generic node or axon-initial-segment structural disruption.","A validated scaffold-to-transcription connection would extend ankyrin-G biology from channel localization to maintenance of the molecular state that supports synaptic and calcium physiology. Because ANK3 is central to neuronal excitability and psychiatric genetics, separating this program from general axonal-structure failure could motivate multiple studies of how membrane scaffolds communicate with the nucleus.","Established work predicted that ankyrin-G organizes axon-initial-segment and nodal membrane domains (https://doi.org/10.1083/jcb.200612012), while SPTAN1 studies defined an ankyrin–spectrin assembly hierarchy (https://www.jneurosci.org/content/37/47/11311) and linked spectrin dysfunction to neurological disease (https://pubmed.ncbi.nlm.nih.gov/36697767/). It did not specifically predict coordinated regulation of Atp2a2, Atp2b1, Erc2, Vgf, and neurotransmitter-receptor transcripts.","In 151 TH Prkcd Grin2c Glut, Ank3 perturbation lowered Ank3 itself (log2FC -0.256, FDR 8.42e-5), Opn3 (-1.053, FDR 9.24e-11), Atp2a2 (-0.360, FDR 9.24e-11), Atp2b1 (-0.337, FDR 3.81e-10), Erc2 (-0.545, FDR 1.75e-8), Vgf (-0.852, FDR 2.40e-8), Ache (-0.688, FDR 2.58e-8), Cacna2d1 (-0.227, FDR 0.00184), Gabra4 (-0.191, FDR 0.00226), Grm1 (-0.214, FDR 0.00770), and Gria4 (-0.205, FDR 0.0167), while increasing Pcsk1n (+0.552, FDR 1.10e-8). Cntn1, Cntnap1, and Sptan1 were each strongly depleted in the same class (respectively log2FC -0.880, -0.822, and -0.529; FDR 4.36e-29, 1.09e-11, and 2.03e-13), yet their Atp2a2, Opn3, Atp2b1, Pcsk1n, and Erc2 rows all had FDR 1.0. The alternative is unequal functional editing, protein turnover, or a perturbation-specific stress response rather than an ankyrin-G-to-nucleus pathway; axon-initial-segment structure and electrical function were not measured, and the positive state was resolved in one high-yield context.","Axon-initial-segment, ankyrin/spectrin, ion-channel, and psychiatric-neurogenetics communities would be most affected. ANK3 and structural comparators should be depleted to matched protein levels, then ANK3 rescued with constructs that retain or lack defined channel/scaffold-binding domains while axon-initial-segment architecture, nascent RNA, and excitability are measured. Domain-selective transcript rescue would distinguish signaling from generic structural injury and connect ANK3 risk alleles to scaffold-dependent transcriptional homeostasis.",22,144,3,4,4
bucket_1_rank_24,distributed_agents,B1,Distributed Agents catalog ranks 1-30,"ADARB1 perturbation recurrently lowers Flnb across cortical, subcortical, and cerebellar neuronal classes","Flnb RNA decreased directionally after Adarb1 perturbation in every tested neuronal class and reached FDR significance in 12 of 22, with large effects across thalamic, midbrain, hypothalamic, Pvalb, cortical, and cerebellar populations. Flnb is therefore a robust neuronal readout of ADAR2 loss, but its interpretation remains open among FLNB editing, another ADAR2-dependent RNA process, and an indirect cytoskeletal-state response.","The result redirects attention from canonical glutamate-receptor substrates toward a recurrent filamin/actin-associated consequence of ADAR2 disruption. If the ADARB1-FLNB connection proves causal, RNA-editing researchers could pursue a new route by which ADAR2 influences neuronal structure, mechanics, or transport, with implications for disease phenotypes not explained by GRIA2 alone.","ADAR2 was already known to edit Q/R sites in FLNA and FLNB, and a study of ADAR2-deficient mouse heart (https://doi.org/10.1371/journal.pone.0200968) examined FLNB editing and expression. That precedent makes a filamin connection biologically plausible, but it did not establish a recurrent neuronal Flnb-abundance decrease across these brain populations.","Adarb1 perturbation lowered Flnb in 151 TH Prkcd Grin2c Glut (log2FC -1.479, FDR 6.67e-44), 155 MB Glut (-1.654, FDR 1.57e-38), 191 MB P MY GABA (-1.477, FDR 7.81e-16), 145 MH-LH TH Glut (-1.198, FDR 6.89e-15), 066 CNU-HYa HY GABA (-0.913, FDR 6.36e-11), and 052 Pvalb Gaba (-0.797, FDR 2.44e-8). Additional significant decreases occurred in 005 L4-5 IT CTX Glut (-0.693, FDR 2.52e-5), 054 CNU-MGE GABA (-0.850, FDR 1.38e-4), 110 CNU-HYa HY MM Glut (-0.912, FDR 1.57e-4), 012 MEA LA CA1 DG Glut (-0.890, FDR 0.0276), and 308 CB GABA (-0.738, FDR 0.0218). Adarb1 itself was depleted in 151 (-0.555, FDR 2.04e-9), 066 (-0.816, FDR 0.0189), and 052 (-0.616, FDR 0.0401). The strongest alternative is an editing-dependent effect on FLNB RNA stability rather than an editing-independent regulatory link; editing sites, FLNB protein, cytoskeletal function, and matched Adar or Gria2 functional engagement were not measured.","RNA-editing, neuronal-cytoskeleton, and ADAR-linked-disease communities would gain a common mechanistic question. Targeted long-read RNA sequencing and site-resolved editing after ADARB1 loss, with wild-type versus catalytically inactive ADAR2 rescue and editing-mimetic FLNB constructs, can separate editing-dependent transcript stability from indirect or editing-independent regulation. The answer would support systematic mapping from ADAR2-dependent structural RNA targets to neuronal morphology and circuit phenotypes.",24,144,5,3,4
bucket_1_rank_25,distributed_agents,B1,Distributed Agents catalog ranks 1-30,MAX perturbation derepresses STAG3 and circadian E-box genes while MGA shares only the STAG3 branch,MAX targeting strongly induced Stag3 in thalamic and midbrain glutamatergic neurons and induced Per2 and Cry1 in the thalamic class; MGA targeting also induced Stag3 and Obscn but shared little whole-transcriptome similarity with MAX. The pattern suggests separable MAX-containing repressor branches in adult neurons—an MGA-associated germline/meiotic branch and a distinct circadian E-box branch—rather than a unitary loss of MYC-driven biosynthesis.,"Confirming separable MAX repressor outputs would revise how the MYC/MAX network is organized in mature neurons and connect chromatin repression of normally silent meiotic genes to neuronal clock control. It would support substantial work on how different MAX partners partition repression, whether inappropriate germline-gene activation harms neurons, and whether clock disruption is mechanistically linked or parallel.","MAX is the shared bHLHZip partner for activating MYC complexes and several repressive MXD/MNT/MGA complexes. Prior work supports MAX/MGA repression of germline genes (https://pubmed.ncbi.nlm.nih.gov/29381691/) and meiotic genes (https://pubmed.ncbi.nlm.nih.gov/27025988/), as well as MAX-dependent repression at circadian E-box promoters (https://pubmed.ncbi.nlm.nih.gov/32225100/). The unexpected feature is that these branches separate within the same neuronal perturbation map rather than moving as one generic MAX program.","In 151 TH Prkcd Grin2c Glut, Max perturbation increased Stag3 (log2FC +3.324, FDR 1.90e-11), Per2 (+1.108, FDR 1.64e-10), and Cry1 (+0.654, FDR 0.0275); in 155 MB Glut it increased Stag3 (+3.047, FDR 5.03e-6). Mga perturbation in 151 increased Stag3 (+3.606, FDR 4.83e-5) and Obscn (+4.435, FDR 1.73e-8). The source found 167 FDR-significant Max gene-by-cell rows across six tested contexts, but MAX and MGA had weak whole-transcriptome similarity. Max RNA itself was not FDR-depleted and had mixed directions, so the result is a Max-locus perturbation phenotype, not demonstrated MAX protein loss. The reported absence of Stag3 induction after MYC/MYCN/E2F-DP targeting is not decisive without matched functional engagement, and direct occupancy or ncPRC1.6 activity was not measured. Concentration of the strongest positive rows in 151/155 is compatible with their high effective power and does not establish restriction to those classes.","MYC/MAX-network, ncPRC1.6, neuronal-chromatin, and circadian-transcription researchers need to determine whether the two branches occupy distinct regulatory complexes. Time-resolved MAX and MGA CUT&RUN with nascent RNA after independently validated editing would test direct, partner-specific control at Stag3 and Per2/Cry1. Separating those branches would motivate substantial work on mature-neuron repressor-complex partitioning, ectopic germline programs, and clock-linked vulnerability.",25,144,5,3,4
bucket_1_rank_27,distributed_agents,B1,Distributed Agents catalog ranks 1-30,GRIN2A raises an L2/3 NPTX2-BDNF-PTGS2 program while GRIN2B trends oppositely and significantly lowers Vgf and Kcnv1,"GRIN2A targeting increased an activity/plasticity-associated RNA set in upper-layer cortical excitatory neurons, whereas GRIN2B targeting in L2/3 moved the same markers downward, with FDR support strongest for Vgf and Kcnv1 on the GRIN2B side. Opposite homeostatic states after chronic loss of GluN2A- and GluN2B-containing receptors are a plausible interpretation, but asymmetry in statistical and on-target support makes subunit non-equivalence provisional.","Validated sign-opposed responses would add a chronic transcriptional dimension to GluN2A/GluN2B specialization and could change how subunit-selective NMDAR interventions are expected to remodel cortical circuits over time. This would seed work connecting receptor composition to homeostatic plasticity, neurotrophin signaling, and long-term drug effects rather than treating both subunits as interchangeable reductions in NMDAR current.","Established work predicted distinct GluN2A and GluN2B disease and circuit roles (https://pubmed.ncbi.nlm.nih.gov/31807283/), while NPTX2/Narp homeostatic synaptic scaling (https://pubmed.ncbi.nlm.nih.gov/20729843/) and NMDA-receptor-dependent neuronal PTGS2 induction (https://pubmed.ncbi.nlm.nih.gov/35354047/) provided marker-level precedents. It did not specifically predict opposite chronic regulation of the Nptx2/Bdnf/Ptgs2/Vgf group in superficial cortical neurons.","In 007 L2-3 IT CTX Glut, Grin2a perturbation increased Nptx2 (log2FC +1.777, FDR 6.09e-4), Bdnf (+1.249, FDR 0.0272), Ptgs2 (+1.264, FDR 0.0353), and Cckbr (+0.832, FDR 0.00117); Vgf also rose (+0.650) but had FDR 0.395. In 005 L4-5 IT CTX Glut, the most secure component was Nptx2 induction (+1.401, FDR 7.94e-6), with Grin2a depletion at -0.349, FDR 0.0239. In 007, Grin2b perturbation depleted Grin2b (-0.471, FDR 0.00107), lowered Kcnv1 (-1.108, FDR 2.89e-5) and Vgf (-0.987, FDR 0.0194), and moved Bdnf (-1.156, FDR 0.184), Cckbr (-0.831, FDR 0.0707), Nptx2 (-1.032, FDR 0.617), and Ptgs2 (-1.453, FDR 0.286) in the opposite direction without FDR support. Thus the multigene reversal rests partly on nominal evidence, and Grin2a itself was not FDR-depleted in L2/3 (log2FC -0.401, FDR 0.274). Unequal editing, receptor turnover, baseline activity, or sampled cell state could explain the contrast; receptor currents and network activity were not measured.","NMDAR-subunit, cortical-plasticity, and neuropsychiatric-pharmacology communities would be most motivated by a verified sign reversal. Independently validated, matched Grin2a and Grin2b edits in identified L2/3 neurons should be followed longitudinally through subunit currents, network activity, and RNA trajectories, with subunit-specific rescue as the causal test. Reversible divergence would create a research direction on chronic subunit-selective homeostasis and delayed responses to NMDAR-modulating drugs.",27,144,4,3,4
bucket_1_rank_28,distributed_agents,B1,Distributed Agents catalog ranks 1-30,"EEF2 targeting induces a large Vgf, Nptx2, and Cx3cl1 trophic state in superficial cortical neurons","EEF2 targeting generated a 575-gene response in L2/3 intratelencephalic cortical neurons, led by strong Vgf, Nptx2, and Cx3cl1 induction and Uchl1 repression, with smaller Vgf- or Ranbp2-positive responses in other neuronal classes. Impaired elongation may provoke a superficial-cortical trophic and neuron-to-glia signaling state, but unchanged Eef2 RNA and the absence of direct translation measurements leave guide-linked or secondary state effects unresolved.","If EEF2 protein disruption causes this state, translational stress in cortical neurons would have an unexpectedly strong secretory and trophic output rather than a simple uniform loss of protein synthesis. That could support major work on how vulnerable excitatory neurons signal adaptation or distress to surrounding cells and whether VGF, NPTX2, or CX3CL1 promotes recovery versus pathology.","EEF2 is the conserved ribosomal translocation factor (https://rest.uniprot.org/uniprotkb/P58252), and neuronal eEF2 phosphorylation is established as a regulator of dendritic translation (https://doi.org/10.3389/fncel.2014.00035). These priors predict translation-related consequences but not a large Vgf/Nptx2/Cx3cl1 abundance state centered in superficial cortical neurons.","In 007 L2-3 IT CTX Glut, Eef2 perturbation increased Vgf (log2FC +2.149, FDR 1.01e-8), Nptx2 (+2.946, FDR 3.02e-8), and Cx3cl1 (+0.864, FDR 4.93e-8), while lowering Uchl1 (-0.712, FDR 5.35e-5); the full response contained 575 FDR-significant genes. Smaller echoes included Vgf induction in 052 Pvalb Gaba (+1.424, FDR 1.35e-5) and Ranbp2 induction in 155 MB Glut (+0.827, FDR 2.60e-5). Eef2 RNA itself was not FDR-depleted in any tested class, and effective protein disruption, elongation rate, ribosome occupancy, cell survival, and secreted peptide levels were not measured. Independent-guide effects, selective survival, or a pre-existing state imbalance are therefore major alternatives, and the stronger L2/3 yield does not by itself establish intrinsic cell-type specificity.","Translation-control, cortical-circuit, neurotrophin, and neuron-glia-signaling researchers would pursue whether this trophic state is adaptive. Independent Eef2 alleles, protein quantification, ribosome profiling or nascent-protein labeling, and time-resolved single-cell RNA would first distinguish elongation stress from guide-linked selection; VGF/CX3CL1 neutralization could then test effects on edited and neighboring cells. Those results would support a program on secretory adaptation to translational stress across cortical layers and disease models.",28,144,4,3,4
bucket_1_rank_29,distributed_agents,B1,Distributed Agents catalog ranks 1-30,"BRAF and MAP2K1 perturbations repress, whereas NF1 perturbation induces, a neuronal VGF-PTPN5-ETV5 MAPK module","In thalamic Prkcd/Grin2c glutamatergic neurons, BRAF and MAP2K1 targeting lowered a shared Vgf/Ptpn5/Unc13a axis, NF1 targeting moved Vgf, Ptpn5, Atp2a2, and Etv5 upward, and SHOC2 targeting also lowered Vgf; NF1-associated Vgf induction recurred in midbrain glutamatergic and midbrain/pons/myelencephalon GABA neurons. These signed RNA responses support a neuronal RAF-MEK branch opposite the NF1 Ras brake, while weaker paralog and LZTR1 comparisons remain boundaries rather than evidence of biochemical dispensability.","The expected pathway polarity becomes valuable here because it is attached to a concrete neuronal output containing VGF, STEP/PTPN5, ETV factors, calcium handling, and release machinery. Validating that output could give RASopathy and neurofibromatosis research a cell-resolved bridge from pathway genotype to neuronal physiology and enable branch-selective biomarkers for interventions.","BRAF is an upstream kinase for MAP2K1 and MAPK activation (https://www.uniprot.org/uniprotkb/P15056/entry). NF1 is a Ras GTPase-activating negative regulator whose loss activates the RAS/RAF/MAPK pathway (https://reactome.org/content/detail/R-HSA-6802957), whereas SHOC2 supports RAF activation dynamics (https://doi.org/10.1073/pnas.1902658116). The expected sign is therefore NF1 loss opposite BRAF/MEK loss; the new element is the specific mature-neuronal RNA module and its recurrence.","In 151 TH Prkcd Grin2c Glut, Map2k1 perturbation lowered Vgf (log2FC -0.785, FDR 7.85e-8), Ptpn5 (-0.436, FDR 9.37e-4), Unc13a (-0.325, FDR 0.0119), and Etv5 (-0.631, FDR 0.0119). Braf perturbation lowered Vgf (-0.795, FDR 1.39e-4), Ptpn5 (-0.635, FDR 1.39e-4), Atp2a2 (-0.263, FDR 1.39e-4), and Unc13a (-0.315, FDR 0.00393). Nf1 perturbation reversed the axis: Vgf +0.848 (FDR 1.36e-11), Ptpn5 +0.550 (FDR 1.64e-8), Atp2a2 +0.260 (FDR 7.30e-5), and Etv5 +0.541 (FDR 0.0127); Shoc2 perturbation lowered Vgf (-0.660, FDR 0.00794). NF1-associated Vgf induction recurred in 155 MB Glut (+2.192, FDR 1.75e-16) and 191 MB P MY GABA (+1.871, FDR 1.78e-9), where Scg2 and Unc13a also rose (+0.856 and +0.482, both FDR 0.00820). Braf RNA depletion was not FDR-significant, Map2k1 self-depletion was not FDR-significant, and direct MAPK phosphorylation or activity was not measured; differential editing, activity-state changes, and Vgf neurotrophin responsiveness could explain part of the polarity. Nominal LZTR1 rows and weak RAF1/MAPK1 negatives do not establish node-specific independence.","RASopathy, neurofibromatosis, neuronal-MAPK, and synaptic-plasticity communities could use this as a candidate functional map of pathway dosage. A matched NF1, BRAF, MAP2K1, and SHOC2 allelic series in identified neuronal classes, with time-resolved phospho-ERK and genetic epistasis, should test whether ERK normalization restores VGF/PTPN5/ETV5 and electrical phenotypes in the predicted order. Causal ordering would support stratification of RAS/MAPK disorders by neuronal branch output and tests of state-selective pathway correction.",29,144,4,3,4
bucket_1_rank_30,distributed_agents,B1,Distributed Agents catalog ranks 1-30,SIN3A targeting induces cilium and cholesterol genes while repressing spliceosome transcripts in thalamic and midbrain neurons,"SIN3A targeting produced a large thalamic response with induction of cilium/centrosome, Hcn3, Cyp46a1, and Grn transcripts alongside repression of Srrm2, Snrnp70, Rnpc3, Ddx17, and other RNA-processing genes; a smaller cilium/cholesterol and spliceosome pattern recurred in midbrain and inhibitory neurons. A SIN3A-dependent chromatin state coordinating ciliary or sterol biology with RNA processing is one interpretation, but the upward Sin3a RNA shift and response breadth leave direct repression, allelic consequence, and causal unity unresolved.","SIN3A-related disease is usually framed through cortical development and intellectual disability. A validated mature-neuronal program joining cilia, brain cholesterol turnover, and spliceosome abundance would broaden that model and create several mechanistically linked directions for Witteveen-Kolk syndrome, synaptic plasticity, and neuronal chromatin research.","SIN3A was expected to act as an HDAC-associated transcriptional scaffold (https://doi.org/10.1016/j.bbagrm.2009.05.007), and haploinsufficiency is established in a neurodevelopmental syndrome with altered cortical development (https://doi.org/10.1038/ng.3619). Postnatal excitatory-neuron deletion also affects hippocampal plasticity through Homer1/mGluR5 signaling (https://doi.org/10.1172/jci.insight.92385). These sources do not predict the observed cilium/cholesterol-plus-spliceosome relationship.","In 151 TH Prkcd Grin2c Glut, Sin3a targeting increased Odf2 (log2FC +0.820, FDR 1.71e-26), Hcn3 (+2.192, FDR 3.99e-23), Dzip1 (+0.840, FDR 4.54e-21), Cyp46a1 (+0.299, FDR 1.30e-10), and Grn (+0.423, FDR 0.00419), while lowering Srrm2 (-0.476, FDR 1.60e-19), Snhg11 (-0.236, FDR 5.84e-19), Snrnp70 (-0.392, FDR 2.39e-11), Rnpc3 (-0.593, FDR 2.76e-11), and Ddx17 (-0.352, FDR 1.05e-9). In 155 MB Glut, the grounded summary reports Cyp46a1 induction with score 7.553 and FDR 1.85e-10, Odf2 induction with score 7.119 and FDR 2.70e-9, Snhg11 repression with score -9.925 and FDR 5.67e-19, and Srrm2 repression with score -7.924 and FDR 1.34e-11. Sin3a RNA itself increased in 151 (+1.354, FDR 2.36e-17) and in four other responsive classes, so its direction cannot be read as measured loss; protein disruption, chromatin occupancy, splicing, ciliary structure, cholesterol flux, and neuronal physiology were not assessed. A broad stress response or guide-specific allele effect could unite the signatures, and weak REST or other chromatin-comparator responses do not prove SIN3A specificity. The dominance of 151/155 may reflect their greater effective power, so it cannot establish preferential biological sensitivity.","SIN3A/Witteveen-Kolk, neuronal-chromatin, cilia, cholesterol, and RNA-processing communities need to determine whether this is one regulatory state or several secondary consequences. Defined Sin3a alleles with SIN3A protein quantification, CUT&RUN/ATAC, and long-read RNA in the same purified neurons would test direct occupancy and splicing, while ciliary imaging and sterol measurements separate downstream branches. Coherent causal ordering would establish a mature-neuron SIN3A program connecting chromatin dosage to organelle, lipid, and RNA-processing phenotypes.",30,144,4,3,4
bucket_2_rank_50,distributed_agents,B2,Distributed Agents catalog ranks 50-59,KMT2A perturbation reveals a mature thalamic Nr2f1-Pcdh7-Lhx9 maintenance state,"Observation: Kmt2a perturbation lowered Nr2f1, Pcdh7, Lhx9, Camk2d, Pcdh10, and Grik2 and increased Igfbp5 and Cx3cl1 in thalamic neurons, while other neuronal classes showed only singletons or a tiny midbrain branch. Interpretation: acute KMT2A loss in mature neurons engages a defined thalamic maintenance regulon rather than a broad developmental HOX or neurogenesis program.",The result extends KMT2A biology from development and memory into a specific mature-neuronal expression state. Its scope is moderate because direct chromatin regulation and physiological consequence remain to be demonstrated.,"Neuronal KMT2A was already required for prefrontal plasticity and working memory (Jakovcevski et al. 2015 (https://pubmed.ncbi.nlm.nih.gov/25834037/)), postnatal neural-stem-cell neurogenesis and Dlx2 activation (Lim et al. 2009 (https://doi.org/10.1038/nature07726)), and human KMT2A mutations cause Wiedemann-Steiner syndrome (Jones et al. 2012 (https://pubmed.ncbi.nlm.nih.gov/22795537/)). The atlas adds a postmitotic thalamic output distinct from those known settings.","In 151 TH Prkcd Grin2c Glut, Kmt2a perturbation lowered Nr2f1 (log2FC -0.806, FDR 2.39e-10), Pcdh7 (-0.774, FDR 3.60e-7), Lhx9 (-1.182, FDR 1.54e-4), and Kmt2a (-0.485, FDR 1.76e-4), while increasing Igfbp5 (+0.874, FDR 0.0126); 24 genes passed FDR<0.05. In 155 MB Glut, Fnbp1l (-0.749, FDR 0.0214) and Kmt2a were the only FDR rows, and three additional glutamatergic classes yielded isolated Ptprk, Aff4, or Nap1l2 hits. Las1l did not phenocopy Kmt2a, but most other 5FMC components were not perturbed, so complex-level exclusion is incomplete. KMT2A protein, H3K4 methylation, occupancy, and neuronal function were not measured.","KMT2A/Wiedemann-Steiner, neuronal chromatin, and thalamic-development researchers would be most interested. Matched acute KMT2A loss with CUT&RUN, H3K4 profiling, nascent RNA, and rescue by catalytic versus scaffold mutants would distinguish direct mature-neuronal targets from secondary state change. Confirmation would support a focused program on postmitotic KMT2A maintenance rather than broad developmental reanalysis.",50,144,3,4,3
bucket_2_rank_51,distributed_agents,B2,Distributed Agents catalog ranks 50-59,YLPM1 perturbation lowers Srsf6-Srsf5 and Inha in midbrain-linked neurons,"Observation: Ylpm1 RNA was reduced across several neuronal classes, but midbrain glutamatergic neurons additionally lowered Srsf6, Srsf5, and Inha, with Inha recurring in posterior GABAergic neurons. Interpretation: YLPM1 loss may stress nuclear-speckle or post-transcriptional machinery in these neurons, although the branch is primarily supported by one high-yield midbrain context.",The result supplies a concrete neuronal output for a poorly characterized nuclear protein and a small set of markers for mechanistic follow-up. It is likely to extend RNA-processing work rather than establish a broad new disease program.,"YLPM1 was annotated as a nuclear protein associated with speckle-related compartments (UniProt YLPM1 (https://www.uniprot.org/uniprotkb/Q9R0I7); Human Protein Atlas (https://www.proteinatlas.org/ENSG00000119596-YLPM1)), and nuclear speckles are defined by RNA-processing factors (QuickGO GO:0016607 (https://www.ebi.ac.uk/QuickGO/term/GO:0016607)). These priors make SRSF changes plausible but do not predict their neuronal context.","In 155 MB Glut, Ylpm1 perturbation lowered Ylpm1 (log2FC -0.652, FDR 6.93e-4), Srsf6 (-0.610, FDR 9.35e-4), Srsf5 (-0.440, FDR 0.00429), and Inha (-0.501, FDR 0.0232), while increasing Shprh (+0.787, FDR 9.35e-4), Jakmip2 (+0.875, FDR 0.0231), and Atxn2l (+0.362, FDR 0.0232). In 191 MB P MY GABA, Ylpm1 and Inha both fell about -0.833 (FDR 5.92e-4). Ylpm1 also fell in 066 CNU-HYa HY GABA (-1.173, FDR 7.09e-4) without a broad companion program. The latter null does not prove state-selective coupling because effective power and protein loss were unmeasured; speckle morphology, splice isoforms, and direct YLPM1 interactions were not assayed.","Nuclear-speckle, splice-regulator, and YLPM1 rare-disease researchers would be the relevant audience. Imaging speckle organization and long-read RNA after matched YLPM1 protein depletion in midbrain and comparator neurons, followed by YLPM1 rescue, would test whether the SRSF changes report genuine speckle dysfunction. Confirmation would support a bounded mechanistic study of YLPM1-dependent RNA processing.",51,144,3,4,3
bucket_2_rank_52,distributed_agents,B2,Distributed Agents catalog ranks 50-59,BMAL1-CLOCK and REV-ERB perturbations recover an asymmetric neuronal circadian feedback loop,"Observation: Bmal1 and Clock perturbations lowered Dbp, Nr1d2, and related clock outputs, whereas Nr1d1 or Nr1d2 perturbations induced portions of the positive clock limb and a synaptic branch in thalamic neurons. Interpretation: the atlas recovers expected circadian feedback but shows asymmetric paralog and endpoint behavior that should be treated as a single-timepoint basal state, not a phase-resolved oscillator.",The finding validates that pooled postnatal perturbation can recover a neuronal clock circuit and identifies a tractable thalamic synaptic output. Its priority is moderate because the central feedback logic is established and the experiment lacks time resolution.,"CLOCK:BMAL1 activation of core clock genes and the PER/CRY and REV-ERB feedback loops are established (Reactome circadian clock (https://reactome.org/content/detail/R-HSA-9909396); Trott and Menet 2018 (https://journals.plos.org/plosgenetics/article?id=10.1371/journal.pgen.1007156)), and REV-ERB paralogs are known to coordinate circadian output (Ikeda et al. 2019 (https://pmc.ncbi.nlm.nih.gov/articles/PMC6629614/)). The atlas primarily adds neuronal context and signed asymmetry.","In 151 TH Prkcd Grin2c Glut, Bmal1 perturbation lowered Dbp (log2FC -1.601, FDR 3.07e-22), Nr1d2 (-1.083, FDR 1.19e-18), and Nr1d1 (-1.074, FDR 4.70e-18), while increasing Erc2 (+0.628, FDR 1.32e-12); 103 genes passed FDR<0.05. Clock perturbation in the same class lowered Dbp -1.030 (FDR 1.36e-11) and Nr1d2 -0.378 (FDR 2.25e-4), and in 155 MB Glut lowered Nr1d2 -0.833 (FDR 1.30e-13) and Dbp -1.290 (FDR 2.05e-6). Nr1d2 perturbation increased Nr1d1 +0.520 in 151 (FDR 1.44e-5), while Nr1d1 perturbation increased Dbp +0.407 (FDR 0.00243) and generated a 116-gene branch. A single harvest time can invert or attenuate feedback depending on phase, and neither circadian period nor rhythmic amplitude was measured.","Circadian-neurobiology and thalamic-circuit researchers would use the result mainly as a starting map. Time-resolved perturbation across circadian phase, with bioluminescent clock reporters and electrophysiology, would distinguish altered oscillator dynamics from static phase displacement. Confirmation would support a focused program on how the neuronal clock gates synaptic-state genes in defined thalamic cells.",52,144,3,4,3
bucket_2_rank_53,distributed_agents,B2,Distributed Agents catalog ranks 50-59,HSPA8 perturbation induces a compact HSP70-HSP90 compensatory transcript arm,"Observation: Hspa8 perturbation increased Dnaja1, Hspa4l, Hsp90aa1, and Cct6a in thalamic neurons, with Dnaja1 or Hsp90aa1 induction recurring in cortical and hypothalamic classes. Interpretation: loss of constitutive HSPA8 releases a compensatory folding response, although correlation with proteasome perturbations does not prove altered folding flux.",The response gives a concise neuronal measure of HSPA8 insufficiency and supports comparison of constitutive versus inducible chaperone networks. It is a well-grounded extension of established feedback biology with moderate experimental value.,"HSPA8 was established as a constitutive HSP70 chaperone (UniProt HSPA8 (https://rest.uniprot.org/uniprotkb/P11142)), DNAJ proteins as HSP70 specificity factors (Kampinga and Craig 2010 (https://doi.org/10.1038/nrm2941)), and HSP70–HSF1 negative feedback as a mechanism for compensatory chaperone induction (Krakowiak et al. 2018 (https://doi.org/10.7554/eLife.31668)). The observed arm therefore has strong precedent.","In 151 TH Prkcd Grin2c Glut, Hspa8 perturbation increased Dnaja1 (log2FC +0.496, FDR 7.23e-4), Hspa4l (+0.394, FDR 7.23e-4), Hsp90aa1 (+0.383, FDR 0.00252), and Cct6a (+0.700, FDR 0.0203). Dnaja1 also increased in 009 L2-3 IT PIR AON ENT Glut (+0.581, FDR 0.0196), and Hsp90aa1 increased in 066 CNU-HYa HY GABA (+0.720, FDR 0.0198). Nominal positive calls for Hsph1, Dnaja1, Dnajb1, Hsp90aa1, Stip1, and Hspa4l recurred across the 21 tested classes. Protein folding, HSF1 activity, aggregation, and proteasome flux were not measured, so the interpretation remains an abundance-level compensatory response rather than demonstrated proteostasis recovery.","Neuronal proteostasis, HSP70/HSP90, and chaperone-therapy researchers would find the marker arm useful. HSF1 reporter assays, folding sensors, proteomics, and wild-type versus ATPase-deficient HSPA8 rescue would determine whether transcript induction restores functional proteostasis. Confirmation would support a focused neuronal chaperone-feedback study rather than a broad new program.",53,144,3,4,3
bucket_2_rank_54,distributed_agents,B2,Distributed Agents catalog ranks 50-59,PPP3CA and PPP3R1 perturbations expose a noncanonical calcineurin transcript output opposite AP-2 adaptor states,"Observation: Ppp3ca and Ppp3r1 perturbations produced compact, partially overlapping thalamic responses centered on Pcsk1n and selected calcium, synaptic, and receptor genes, while AP-2 adaptor perturbations moved the restricted Ppp3r1 signature oppositely. Interpretation: calcineurin loss is captured here by an endocytic and excitability-related RNA module rather than a canonical NFAT/RCAN abundance reporter.","The response suggests a tractable transcriptional interface between calcineurin and synaptic endocytosis, but it does not yet establish phosphatase-to-AP-2 epistasis. It is a moderate-priority refinement of well-known calcineurin biology.",Calcineurin was already established as a calcium-dependent phosphatase that activates NFAT (Reactome (https://reactome.org/content/detail/R-HSA-2025928)) and dephosphorylates endocytic proteins during synaptic-vesicle recycling (Cousin and Robinson 2001 (https://pubmed.ncbi.nlm.nih.gov/11672811/)). Those priors make both NFAT and endocytic outputs plausible; the atlas instead emphasizes a compact non-NFAT RNA state.,"In 151 TH Prkcd Grin2c Glut, Ppp3ca perturbation lowered Ppp3ca (log2FC -0.483, FDR 6.35e-12) and Iyd (-1.085, FDR 0.0112) while changing Ptpn3 (-0.224, FDR 0.0239) and Pik3r1 (+0.488, FDR 0.0239). Ppp3r1 perturbation lowered Ppp3r1 (-0.703, FDR 1.89e-5), Atp2a2 (-0.327, FDR 8.08e-5), Coro6 (-0.472, FDR 5.99e-4), and Slc6a17 (-0.286, FDR 0.0496), while increasing Pcsk1n (+0.710, FDR 2.21e-7) and Ptprs (+0.356, FDR 4.87e-7). Ap2s1 and Ap2m1 were the strongest negative signature neighbors; for example, Ap2s1 perturbation increased Atp2a2 and Slc6a17 while lowering Ptprs and Ndrg2. That anticorrelation is not double-perturbation epistasis, and NFAT localization, phosphatase activity, endocytosis, and calcium were not measured.","Calcineurin, synaptic-vesicle endocytosis, and calcium-signaling researchers would be most interested. Matched calcineurin-subunit and AP-2 perturbations with phosphatase assays, endocytic imaging, NFAT localization, and double-perturbation rescue would distinguish a causal countermodule from correlated neuronal states. Confirmation would support a focused study of calcineurin–endocytosis coupling in mature thalamic neurons.",54,144,3,4,3
bucket_2_rank_55,distributed_agents,B2,Distributed Agents catalog ranks 50-59,ACTB perturbation recurrently induces ACTG1 and FLNA without reciprocal ACTG1 compensation,"Observation: Actb perturbation increased Actg1 and Flna across multiple neuronal classes, whereas Actg1 perturbation did not produce a broad reciprocal Actb-like state and yielded only a narrow Flna response. Interpretation: cytoplasmic actin compensation is directionally asymmetric and includes a filamin-associated remodeling arm.","The pattern gives neuronal context to established β- versus γ-actin nonredundancy and could help interpret cytoskeletal disease alleles. It is robust but closely aligned with known actin biology, so follow-up should remain proportionate.","β- and γ-cytoplasmic actins were already known to encode nearly identical proteins with nonredundant gene-level functions (Vedula et al. 2017 (https://pubmed.ncbi.nlm.nih.gov/29244021/); Patrinostro et al. 2018 (https://pubmed.ncbi.nlm.nih.gov/30012594/)), and ACTB/ACTG1 variants cause a shared developmental syndrome (Rivière et al. 2012 (https://pubmed.ncbi.nlm.nih.gov/22366783/)). Asymmetric compensation was therefore plausible.","After Actb perturbation, Actg1 increased in 9 of 22 contexts and Flna in 13 of 22 at FDR<0.05. In 151 TH Prkcd Grin2c Glut, Actg1 rose log2FC +0.695 (FDR 1.32e-16) and Flna +0.802 (FDR 2.54e-13); in 027 NP-CT-L6b-OB Glut, Actg1 rose +0.651 (FDR 0.00376) and Flna +1.415 (FDR 7.93e-15); in 155 MB Glut, the changes were +0.869 (FDR 3.33e-19) and +0.669 (FDR 0.0212). Direct Actg1 perturbation strongly lowered Actg1 in 151 (-0.824, FDR 9.85e-16) but did not significantly induce Flna there; its only non-self FDR hit was Flna in 027 (+0.779, FDR 0.0132). Actb itself was absent from response rows, and mRNA does not measure actin protein stoichiometry, polymerization, or morphology.","Actin-autoregulation, neuronal-cytoskeleton, migration, and Baraitser-Winter researchers would find the asymmetry useful. Isotype-specific protein quantification, polymerization imaging, and reciprocal coding-sequence/UTR rescue after defined Actb and Actg1 alleles would distinguish locus regulation from protein demand. Confirmation would support a focused neuronal study of actin-isotype compensation.",55,144,3,4,3
bucket_2_rank_56,distributed_agents,B2,Distributed Agents catalog ranks 50-59,"CAML perturbation lowers receptor, ion-pump, and synaptic-vesicle transcripts","Observation: Caml perturbation lowered receptor and membrane genes in thalamic neurons and vesicle, chloride-transporter, ion-pump, and receptor genes in midbrain glutamatergic neurons. Interpretation: loss of the WRB–CAML tail-anchored-protein insertion machinery creates a neuronal membrane-protein liability reflected at the RNA level, although insertion and localization were not measured.",The response extends established ER insertion biology into mature neuronal receptor and vesicle maintenance and supplies concrete readouts for mechanistic work. Its expected contribution is moderate because the pathway itself is known and the transcript changes could be secondary compensation.,"CAML/CAMLG was already part of the mammalian WRB receptor for ER insertion of tail-anchored proteins (Yamamoto and Sakisaka 2012 (https://pubmed.ncbi.nlm.nih.gov/23041287/); Vilardi et al. 2014 (https://pubmed.ncbi.nlm.nih.gov/24392163/); Reactome WRB–CAMLG docking (https://reactome.org/content/detail/R-HSA-9609899)). A membrane-protein response was plausible, but the neuronal receptor and vesicle transcript set was not directly predicted.","In 151 TH Prkcd Grin2c Glut, Caml perturbation lowered Lingo1 (log2FC -0.327, FDR 3.88e-4), Gabrd (-0.281, FDR 0.00239), Gabra1 (-0.286, FDR 0.0134), Syp (-0.207, FDR 0.0154), and Caml (-0.375, FDR 0.0285); 28 genes passed FDR. In 155 MB Glut, it lowered Sv2a (-0.420, FDR 0.00303), Caml (-0.553, FDR 0.00336), Syp (-0.246, FDR 0.00669), Slc12a5 (-0.397, FDR 0.00849), Atp1a3 (-0.227, FDR 0.00857), and Gabbr1 (-0.295, FDR 0.0208); 18 genes passed FDR. Most other contexts had no corrected downstream response, but those nulls do not establish intrinsic restriction. Direct tail-anchor insertion, receptor surface abundance, ER stress, and synaptic function were not assayed.","Tail-anchored-protein insertion, synaptic-membrane trafficking, and neuronal proteostasis researchers would be the main audience. Surface proteomics and insertion reporters after matched CAML and WRB loss, with CAML rescue, would determine whether the RNA state follows failed insertion of specific neuronal cargos. Confirmation would support a focused program on how ER insertion capacity maintains synaptic membrane composition.",56,144,3,4,3
bucket_2_rank_57,distributed_agents,B2,Distributed Agents catalog ranks 50-59,SRSF1 perturbation rewires splice-regulator abundance instead of inducing SRSF2 compensation,"Observation: Srsf1 perturbation increased Ptbp2, Rbm28, Dazap1, and Psip1 while lowering Srsf2, Hnrnpdl, Clk4, and Clk1 in responsive neurons. Interpretation: SRSF1 loss destabilizes a broader RNA-regulator network rather than eliciting simple SRSF2 or CLK-family transcriptional compensation.",The response identifies a concrete abundance-level network reaction to loss of a core SR protein and can guide isoform-focused experiments. It is a moderate-priority extension because direct splicing outcomes and functional compensation remain unknown.,"SRSF1 was already established as an essential SR-family splicing regulator (UniProt SRSF1 (https://rest.uniprot.org/uniprotkb/Q6PDM2)), and interaction/enrichment resources place the response genes in a shared RNA-processing neighborhood (STRING (https://string-db.org/)). The signed collapse of Srsf2 and CLK transcripts rather than compensatory induction was not guaranteed by those priors.","In 151 TH Prkcd Grin2c Glut, Srsf1 perturbation strongly lowered Srsf1 (log2FC -1.832, FDR 7.05e-65) and Srsf2 (-1.222, FDR 1.09e-44), while increasing Ptbp2 (+1.194, FDR 2.99e-72), Rbm28 (+1.248, FDR 4.09e-67), Dazap1 (+1.260, FDR 9.95e-65), and Psip1 (+0.999, FDR 4.94e-64). Related, uneven RNA-processing responses occurred in CA3/CA2/DG glutamatergic and hypothalamic or striatal GABAergic contexts. The very broad thalamic response may include secondary stress, and total RNA abundance neither measures splice isoforms nor demonstrates loss of functional SRSF2 compensation.","Spliceosome, SR-protein, neuronal RNA-processing, and RBP-disease researchers would be most interested. Long-read RNA-seq, phospho-SR proteomics, and SRSF2 or CLK rescue after matched SRSF1 depletion would distinguish failed compensation from downstream network remodeling. Confirmation would support a focused study of SR-network homeostasis in neurons.",57,144,3,4,3
bucket_2_rank_58,distributed_agents,B2,Distributed Agents catalog ranks 50-59,HSF1 perturbation lowers a basal HSPH1-HSP90 support module,"Observation: Hsf1 perturbation lowered Hsph1, Hsp90aa1, Hsp90ab1, Ahsa2, Stip1, and Rsrp1 in selected neuronal classes. Interpretation: HSF1 maintains a compact constitutive chaperone support layer in mature neurons, distinct from the sparse response to direct Hsp90ab1 loss.","The result provides cell-resolved evidence for basal, not merely heat-induced, HSF1 control of neuronal chaperone abundance. It is a useful validation and refinement of established feedback biology with moderate follow-up value.","HSF1 was already the central heat-shock transcription factor (Reactome HSF1 activation (https://reactome.org/content/detail/R-HSA-3371511); Gomez-Pastor et al. 2018 (https://pubmed.ncbi.nlm.nih.gov/28852220/)), and HSP90 regulation was well characterized (Prodromou 2016 (https://pmc.ncbi.nlm.nih.gov/articles/PMC4980810/)). Basal maintenance of a compact chaperone module was therefore plausible.","In 151 TH Prkcd Grin2c Glut, Hsf1 perturbation lowered Hsph1 (log2FC -0.484, FDR 1.69e-8), Hsp90aa1 (-0.282, FDR 2.85e-4), Rsrp1 (-0.175, FDR 8.06e-4), Hsp90ab1 (-0.345, FDR 8.96e-4), Mat2a (-0.234, FDR 0.00737), and Ahsa2 (-0.294, FDR 0.00827). In 155 MB Glut, Hsph1 fell -0.599 (FDR 2.50e-4), Rsrp1 -0.161 (FDR 0.00169), and Stip1 -0.388 (FDR 0.00583); 052 Pvalb Gaba lowered Rsrp1 and Ahsa2, and 110 CNU-HYa HY MM Glut lowered Hsp90aa1 -0.439 (FDR 0.0275). Hsf1 self RNA never passed FDR despite predominantly negative trends, which is non-diagnostic for functional editing. Protein abundance, heat-shock inducibility, aggregation, and survival were not measured.","Heat-shock, neuronal proteostasis, and chaperone-therapy researchers would be the natural audience. Matched HSF1 depletion with chromatin occupancy, basal and induced chaperone proteomics, heat-stress survival, and HSF1 rescue would test direct maintenance of the module. Confirmation would support a focused study of basal HSF1 dependence across neuronal states.",58,144,3,4,3
bucket_2_rank_59,distributed_agents,B2,Distributed Agents catalog ranks 50-59,CTCF perturbation robustly represses clustered protocadherins in thalamic and midbrain neurons,"Observation: Ctcf perturbation coordinately lowered multiple α-, β-, and γ-clustered protocadherins in thalamic and midbrain glutamatergic neurons, alongside a recurrent Zcwpw2-up and Dennd6b/Abhd14b-down branch. Interpretation: the atlas recovers established CTCF control of clustered protocadherin loci and adds a cell-resolved mature-neuronal abundance profile.","This is a strong positive control for relational recovery and a useful map of adult neuronal response, but the central CTCF–protocadherin relationship is already established. Follow-up should focus on the less-explained companion branch and cell-state consequences rather than present a new locus mechanism.","Neuronal CTCF was already required for stochastic clustered-protocadherin expression, dendritic development, and synapse formation (Hirayama et al. 2012 (https://pubmed.ncbi.nlm.nih.gov/22854024/)), and for experience-dependent neuronal gene regulation and chromatin structure (Sams et al. 2016 (https://pubmed.ncbi.nlm.nih.gov/27880914/)). Coordinated Pcdh-cluster repression was therefore expected.","In 151 TH Prkcd Grin2c Glut, Ctcf perturbation lowered Pcdhb20 (log2FC -1.521, FDR 1.47e-7), Pcdhb17 (-1.618, FDR 2.53e-7), Pcdhb16 (-1.910, FDR 2.55e-7), Pcdhac1 (-1.386, FDR 3.28e-7), Pcdhb18 (-1.402, FDR 2.08e-5), and Pcdha10 (-1.106, FDR 1.00e-4). In 155 MB Glut, it lowered Pcdhgb2 (-2.114, FDR 4.17e-5), Pcdhga12 (-1.745, FDR 7.36e-5), Pcdhb20 (-1.933, FDR 7.36e-5), Pcdhb18 (-2.478, FDR 3.51e-5), and Pcdha10 (-1.726, FDR 2.82e-5). The same two classes also showed Zcwpw2 induction and Dennd6b/Abhd14b repression. The assay did not measure CTCF occupancy, chromatin loops, or single-cell protocadherin choice, and the companion genes could reflect a broader state change.","Chromatin-architecture, protocadherin, and neuronal-identity researchers would use this chiefly as an in vivo perturbation map. CTCF CUT&RUN, locus-conformation assays, and single-cell long-read protocadherin profiling after matched loss would verify whether abundance changes reflect altered promoter choice and looping. The proportionate follow-up is to define the unexplained Zcwpw2/Dennd6b/Abhd14b branch and its relationship to the known locus phenotype.",59,144,2,4,3
bucket_3_rank_120,distributed_agents,B3,Distributed Agents catalog ranks 120-129,Sec63 perturbation induces an ER-chaperone response distinct from selected secretory-pathway perturbations,"Sec63 perturbation increased Hspa5, Canx, Hyou1, and Chac1 in 151 TH Prkcd Grin2c Glut, with an Hspa5 echo in 155 MB Glut. Perturbations of Ssr4, Stt3b, Sec23a, Sar1b, Ergic1, and Edem3 showed different expression profiles from this chaperone combination. The pattern defines a SEC63-associated ER import and folding-stress response for comparison with neighboring secretory-pathway perturbations.","The compact response offers a neuronal assay for SEC63 insufficiency and separates it from several neighboring secretory-pathway perturbations at the RNA endpoint. It is a useful mechanistic refinement of established ER-import biology, with limited broader impact until client proteins and functional consequences are identified.","SEC63 was expected to assist SEC61-mediated ER import based on mammalian transport experiments by Lang and colleagues (https://pubmed.ncbi.nlm.nih.gov/22375059/), and its disease relevance includes impaired polycystin biogenesis in genetic interaction studies (https://pubmed.ncbi.nlm.nih.gov/21685914/). CHAC1 is a known ATF4-ATF3-CHOP output from UPR studies (https://pubmed.ncbi.nlm.nih.gov/19109178/), so chaperone and Chac1 induction were consistent with import stress.","In 151 TH Prkcd Grin2c Glut after Sec63 perturbation, Hspa5 increased by log2FC +0.924 (FDR 6.31e-4), Canx by +0.723 (FDR 0.00607), Hyou1 by +0.633 (FDR 0.00876), and Chac1 by +2.030 (FDR 0.0179). Hspa5 also increased in 155 MB Glut by +1.028 (FDR 0.0298). In 151, Ssr4, Stt3b, and Ergic1 each showed strong FDR-supported self-transcript depletion (-0.640, FDR 5.65e-5; -0.619, FDR 3.69e-4; and -0.820, FDR 2.56e-17, respectively) without the reported Sec63 chaperone module, but equivalent protein-level disruption and client load were not measured. Sec63 itself lacked an FDR self-row in the available evidence.","ER-translocation and SEC63-disease researchers could determine whether the response reflects failure of a restricted client set. Confirmed SEC63 editing should be followed by translocation reporters with diverse signal peptides, ER proteomics, BiP occupancy, and rescue by wild-type versus J-domain-defective SEC63, alongside matched Ssr4 and Stt3b controls. Identification of the clients whose import failure precedes the chaperone state would support a focused neuronal SEC63 mechanism.",120,144,3,3,3
bucket_3_rank_121,distributed_agents,B3,Distributed Agents catalog ranks 120-129,"ADNP perturbation defines a CHD4-adjacent neuronal signature across thalamic, midbrain, caudal GABAergic, and Pvalb populations","ADNP perturbation produced a distributed response comprising 25 FDR genes in 151 TH Prkcd Grin2c Glut, three in 155 MB Glut, two in 191 MB P MY GABA, and one in 052 Pvalb Gaba, with recurrent Cacnb3 induction and Plppr3 or Atp9a repression. A report-level signature comparison placed CHD4 nearest to ADNP in 151 and 155. This marker pattern defines a mature-neuronal ADNP output aligned with the established ADNP–CHD4 ChAHP relationship.","If reproduced, the result would provide a compact set of neuronal readouts for testing an established ADNP–CHD4 complex in postnatal neurons. Its value is mainly to extend known ChAHP biology into several mature neuronal populations and to identify tractable markers, consistent with moderate rather than program-defining priority.","Human genetics established ADNP as an autism and intellectual-disability gene (https://pubmed.ncbi.nlm.nih.gov/24531329/), and biochemical work placed ADNP and CHD4 together in the ChAHP complex (https://pubmed.ncbi.nlm.nih.gov/29795351/). Some convergence was therefore expected; the sparse, marker-level neuronal output is the atlas-specific extension.","After Adnp perturbation, Xylt2 decreased by log2FC -0.925 (FDR 1.54e-10), Cpne7 decreased by -0.906 (FDR 1.81e-9), and Vstm2l increased by +1.583 (FDR 7.76e-6) in 151 TH Prkcd Grin2c Glut. Cacnb3 increased in 052 Pvalb Gaba (+1.093, FDR 2.70e-5) and 155 MB Glut (+0.863, FDR 1.23e-4), while Plppr3 decreased in 191 MB P MY GABA (-1.054, FDR 0.0335). The CHD4 link comes from signature proximity rather than chromatin occupancy, and the small, nonidentical marker sets leave generic perturbation-state or indirect transcriptional effects as alternatives.","ADNP syndrome, ChAHP, and neuronal chromatin communities could use these markers to ask whether an established complex has a distinct postnatal role. A decisive test would combine independently validated ADNP editing with CHD4 occupancy or CUT&Tag at the implicated loci and ADNP rescue in the same neuronal populations. Confirmation would justify a focused study of mature-neuron ChAHP targets, not yet a broad new program.",121,144,3,3,3
bucket_3_rank_122,distributed_agents,B3,Distributed Agents catalog ranks 120-129,"PPP2R1A perturbation defines a cortical Amy1-high, Rsrp1/Leng8-low reporter state with TSC/STXBP1 neighborhood structure","In 005 L4-5 IT CTX Glut, PPP2R1A perturbation induced Amy1, Rnpc3, Stmn4, Jun, Pcsk1n, Ovgp1, and related transcripts while repressing Rsrp1, Leng8, Kifc2, and Pabpn1. Signature comparisons placed this cortical marker panel near TSC/STXBP1-positive and LIMK1/DNAJB6-negative states. The resulting reporter profile provides an atlas-defined phenotype for testing how PP2A scaffold perturbation relates to these cortical signaling contexts.","The panel could become a useful RNA phenotype for testing how PP2A scaffold disruption affects mature cortical neurons, including neurodevelopmental-disease alleles. At present it is more likely to support a bounded follow-up study than change PP2A pathway models.","PP2A is organized from scaffold, catalytic, and regulatory subunits (https://pubmed.ncbi.nlm.nih.gov/17174897/), and PPP2R1A or PPP2R5D variants cause PP2A-related intellectual disability (https://pubmed.ncbi.nlm.nih.gov/26168268/). A neuronal transcriptional consequence was plausible, but this exact Amy1/RNA-processing marker combination was not an established expectation.","PPP2R1A perturbation in 005 L4-5 IT CTX Glut increased Amy1 by log2FC +1.181 (FDR 3.02e-18), Rnpc3 by +0.676 (FDR 2.70e-14), and Stmn4 by +0.475 (FDR 8.33e-7), while decreasing Rsrp1 by -0.379 (FDR 1.16e-11), Leng8 by -0.359 (FDR 3.53e-9), and Kifc2 by -0.379 (FDR 3.35e-8). These rows firmly support a reporter state, but the neighbor correlations were report-derived and no phosphatase activity, PP2A holoenzyme composition, or TSC/STXBP1 epistasis was measured.",PP2A neurodevelopmental-disease and cortical signaling researchers would be the most direct audience. Isogenic PPP2R1A editing followed by phosphatase-complex profiling and rescue with wild-type versus disease alleles would distinguish a PP2A-dependent state from guide or stress effects; targeted TSC1/STXBP1 epistasis could then test the tentative neighborhood. A positive result would yield a practical assay for allele stratification.,122,144,3,3,3
bucket_3_rank_123,distributed_agents,B3,Distributed Agents catalog ranks 120-129,Distinct endomembrane perturbations converge on a proteostasis-associated neighborhood with node-specific outputs,"The atlas places ATP6V0C, CAML, PI4KA, DHDDS, VPS53, and VPS54 perturbations near endomembrane or proteostasis states, with node-specific leading outputs: ATP6V0C lowers Atp6v0a1, CAML raises Stmn4, and PI4KA lowers Slc12a5. These results define a many-to-one neighborhood in which acidification, membrane insertion, phosphoinositide synthesis, glycosylation, and tethering converge on related RNA states. The shared neighborhood and its node-specific outputs provide a framework for separating common organelle responses from pathway-specific effects.","This view could help endomembrane researchers select shared stress readouts while preserving the distinct initiating lesions. Because the group is heterogeneous and supported partly by modest correlations, it currently refines experimental organization more than it establishes a new pathway.","ATP6V0C is a V-ATPase V0 pore component (https://www.uniprot.org/uniprotkb/P63082), PI4KA maintains plasma-membrane PI4P and PI(4,5)P2 pools (https://pubmed.ncbi.nlm.nih.gov/24415756/), DHDDS supplies dolichyl-phosphate for glycosylation (https://reactome.org/content/detail/R-HSA-4419978), and VPS53 participates in endocytic recycling (https://pubmed.ncbi.nlm.nih.gov/25799061/). Convergence at a broad endomembrane-stress level was therefore plausible, whereas a single shared neuronal mechanism was not.","ATP6V0C perturbation reduced Atp6v0a1 in 151 TH Prkcd Grin2c Glut (log2FC -0.577, FDR 2.39e-19), 155 MB Glut (-0.438, FDR 6.82e-13), and 052 Pvalb Gaba (-0.450, FDR 1.29e-7). CAML perturbation increased Stmn4 in 191 MB P MY GABA (+1.068, FDR 1.83e-4) and 052 Pvalb Gaba (+0.833, FDR 0.00363); PI4KA perturbation reduced Slc12a5 in 191 MB P MY GABA (-0.670, FDR 0.0136). The broader neighborhood rests on correlation and low-count marker findings, and the PI4KA pineal branch is low-n; shared RNA proximity cannot establish common flux, organelle physiology, or causal ordering.","Neuronal membrane-trafficking, lysosome, and proteostasis groups could test whether the shared neighborhood reflects a common organelle state. Parallel perturbations with live endolysosomal pH, PI4P, glycosylation, and cargo-trafficking measurements would determine whether the RNA overlap tracks one physiological variable or several. The proportionate outcome would be a cross-node phenotyping framework and a small number of mechanistically coherent subgroups.",123,144,3,3,3
bucket_3_rank_124,distributed_agents,B3,Distributed Agents catalog ranks 120-129,KLF9 perturbation defines a Wsb1/Rasgrp2/Pld3 thalamic module distinct from KLF7 and KLF13 outputs,"Klf9-assigned cells in 151 TH Prkcd Grin2c Glut showed 19 FDR genes, including increased Wsb1, Rasgrp2, Pld3, Dzip1, Maged2, Hcn3, and Aplp1. Klf7 and Klf13 perturbations showed different leading marker directions in the same context. Together, these results define a KLF9-associated mature-thalamic state that distinguishes this perturbation from the selected KLF paralogs.",The result could identify a few downstream markers for comparing hormone-responsive KLF paralogs in mature neurons. Its mechanism and consequence are presently too narrow for more than a low-priority validation study.,"Klf9 is directly induced by thyroid hormone in the postnatal brain (https://pubmed.ncbi.nlm.nih.gov/19359381/) and is required for late neuronal maturation (https://pubmed.ncbi.nlm.nih.gov/19657039/). KLF paralogs can differ in their effects on CNS axon growth (https://pubmed.ncbi.nlm.nih.gov/19815778/), so paralog non-equivalence was plausible even though this marker set was unexpected.","Under Klf9 guides in 151 TH Prkcd Grin2c Glut, Wsb1 rose by log2FC +0.316 (FDR 1.31e-7), Rasgrp2 by +0.759 (FDR 2.16e-6), Pld3 by +0.560 (FDR 5.60e-6), Dzip1 by +0.414 (FDR 1.34e-5), Hcn3 by +0.921 (FDR 3.88e-4), and Aplp1 by +0.291 (FDR 0.00533). Klf9 itself changed by -0.138 with FDR 1.0. In Klf7 and Klf13 perturbations, Wsb1, Rasgrp2, and Pld3 were not FDR-supported; however, unequal functional editing could still explain the apparent paralog boundary, and the weak SIN3A similarity does not establish a corepressor mechanism.","Thyroid-hormone, stress-hormone, and neuronal KLF researchers could validate the marker set with independent Klf9 guides, protein-level disruption, and rescue. KLF9 occupancy at Wsb1, Rasgrp2, and Pld3 would distinguish direct regulation from a secondary state. If confirmed, this would support a focused mature-neuron KLF9 regulon study.",124,144,4,3,2
bucket_3_rank_125,distributed_agents,B3,Distributed Agents catalog ranks 120-129,ZMYM2 perturbation induces a Notum/Echdc2/F11r/Atp2a3 thalamic signature,"Zmym2 guides yielded 22 FDR genes in 151 TH Prkcd Grin2c Glut, with strong induction of Echdc2, F11r, Atp2a3, Ebf4, and Notum. This marker state links ZMYM2 perturbation to Notum signaling and Atp2a3-associated calcium handling in mature thalamic neurons. The implicated genes provide defined targets for testing ZMYM2-dependent chromatin regulation.","The signature offers a small set of candidate readouts for studying ZMYM2 in mature neurons and testing whether Notum or calcium handling is affected. Without recurrence, occupancy, or functional consequence, it remains an incremental, low-priority extension of ZMYM2 chromatin biology.","ZMYM2 participates in chromatin corepressor complexes (https://elifesciences.org/articles/86669), and heterozygous loss-of-function variants cause a syndromic congenital disorder (https://doi.org/10.1016/j.ajhg.2020.08.013). Those priors predicted transcriptional effects, but not this specific Notum/Atp2a3 combination.","In 151 TH Prkcd Grin2c Glut, Zmym2 perturbation increased Echdc2 by log2FC +1.343 (FDR 1.30e-11), F11r by +1.069 (FDR 4.31e-10), Atp2a3 by +2.262 (FDR 7.15e-10), Ebf4 by +1.346 (FDR 3.58e-8), and Notum by +0.933 (FDR 4.10e-5). The measured Zmym2 transcript was not FDR-reduced in any class, and no supplied evidence connects ZMYM2 directly to these genes; guide-specific effects or an indirect state change remain substantial alternatives.","ZMYM2 syndrome and chromatin-corepressor groups could perform independent editing and wild-type rescue in the same neuronal class, followed by ZMYM2 occupancy and calcium-store assays centered on Atp2a3. Confirmation would justify a narrow study of this mature-neuronal output rather than a broad disease program.",125,144,4,3,2
bucket_3_rank_126,distributed_agents,B3,Distributed Agents catalog ranks 120-129,SP4 perturbation defines a Tsga10/Abhd11/Svip module in posterior glutamatergic neurons,"The atlas reports corrected responses for SP4 perturbation in 151 TH Prkcd Grin2c Glut and 155 MB Glut, with Tsga10 and Abhd11 induction and supporting Svip, Strada, and Dnajc4 changes. This posterior glutamatergic marker program is distinct from developmental SP4–NMDAR biology. It provides an adult-neuron context for testing SP4 regulation of Tsga10, Abhd11, and the associated genes.","Tsga10 and Abhd11 could be useful readouts for testing an adult SP4 state, but the current evidence does not assign a function to the module. The necessary first step is data-label and perturbation validation, making this a low-priority lead.","Curated annotation describes SP4 as a nuclear GC/GT-box-binding transcription factor (https://www.uniprot.org/uniprotkb/Q02446/entry), while brain-expression resources show broad rather than sharply regional SP4 expression (https://www.proteinatlas.org/ENSG00000105866-SP4). A posterior Tsga10/Abhd11 state was not an established expectation.","The detailed rows label the corrected marker set as 155 MB Glut: Svip log2FC +0.643 (FDR 3.17e-6), Strada +0.469 (FDR 2.23e-5), Tsga10 +0.772 (FDR 7.23e-4), Dnajc4 +0.608 (FDR 0.0146), and Abhd11 +0.529 (FDR 0.0298). A separate evidence summary assigns a strong signature to 151 and a second signature to 155, but its quoted values do not match those detailed cell labels. That inconsistency, together with increased rather than decreased Sp4 RNA, prevents a stronger biological conclusion.","SP-family and adult neuronal transcription researchers should first reproduce the contrast from the source matrix and verify cell labels, then use independent SP4 editing, protein measurement, and rescue. SP4 occupancy at Tsga10 and Abhd11 would distinguish direct transcriptional control from a guide-associated state. Until then, only a small validation project is warranted.",126,144,4,3,2
bucket_3_rank_127,distributed_agents,B3,Distributed Agents catalog ranks 120-129,ZFP148 perturbation lowers spliceosomal and synaptic-marker transcripts in thalamic Prkcd/Grin2c neurons,"Zfp148 perturbation produced its corrected response in 151 TH Prkcd Grin2c Glut, where Srrm2, Srsf9, Cacnb3, Slitrk5, Ptprn, and Rasgrp2 decreased. The resulting marker state connects ZFP148 perturbation with spliceosome-associated and synaptic transcripts. Comparisons with TRP53, HDAC8, CELF2, and ESRRA position it as a distinct atlas-defined state for testing transcriptional and splicing consequences.","The finding provides a testable entry point into an understudied neuronal output of ZNF148. Because the program is small and no altered isoforms or direct targets were measured, its likely near-term value is a focused mechanistic note rather than a broad research direction.","Reactome places SRSF9, SRRM2, and LUC7L2 in mRNA-splicing machinery (https://reactome.org/content/detail/R-HSA-72163), while ZNF148 paralog cooperation is known to be context dependent (https://pubmed.ncbi.nlm.nih.gov/37948511/). These priors make the marker identities interpretable but do not predict direct ZFP148 control of them.","Within 151 TH Prkcd Grin2c Glut, Zfp148 perturbation decreased Srrm2 by log2FC -0.288 (FDR 0.00164), Cacnb3 by -0.939 (FDR 0.00255), Srsf9 by -0.569 (FDR 0.0103), Slitrk5 by -0.762 (FDR 0.0295), and Ptprn and Rasgrp2 by -0.350 and -0.948 (both FDR 0.0450). Luc7l2 was just outside threshold (-0.249, FDR 0.0543), and Zfp148 itself was unchanged (-0.044, FDR 1.0). The major gap is that abundance changes in splice factors do not demonstrate altered splicing or direct ZFP148 regulation.","Neuronal transcription and RNA-splicing groups could test independently edited Zfp148 neurons by long-read or junction-level RNA sequencing, coupled to ZFP148 occupancy and rescue. A reproducible isoform consequence would turn the marker panel into a defined ZFP148–splicing study; without it, the result should remain a small observational lead.",127,144,4,3,2
bucket_3_rank_128,distributed_agents,B3,Distributed Agents catalog ranks 120-129,NR2F1 perturbation couples Sorl1 induction to Sema7a repression in thalamic Prkcd/Grin2c neurons,"Nr2f1 perturbation increased Sorl1 and decreased Sema7a at FDR significance in 151 TH Prkcd Grin2c Glut. Cortical L5-6 IT and CTX-CGE GABA populations exhibited different downstream profiles. Together, the results define an adult NR2F1-associated thalamic response centered on opposing membrane-trafficking and guidance genes.",The opposing Sorl1/Sema7a pair is easy to retest and may connect an established developmental transcription factor to mature neuronal membrane or guidance biology. Two genes and one responsive population limit its expected impact to a low-priority follow-up.,"NR2F1 is an established neurodevelopmental transcription factor associated with Bosch-Boonstra-Schaaf optic atrophy syndrome (https://www.ncbi.nlm.nih.gov/books/NBK587319/), and earlier work established important neural functions for COUP-TF family dosage (https://pubmed.ncbi.nlm.nih.gov/18165280/). A sparse adult Sorl1/Sema7a response was not the obvious prediction.","In 151 TH Prkcd Grin2c Glut, Nr2f1 perturbation increased Sorl1 by log2FC +0.488 (FDR 0.00994) and decreased Sema7a by -0.350 (FDR 0.0407). Nr2f1 itself was only nominally lower (-0.225, FDR 1.0); Chst1, Igsf21, Lrrtm4, Ncan, and Plxdc1 were also nominal rather than FDR-supported. The two-gene output may be indirect, and silence in other populations is not evidence of thalamic specificity under unequal power.","NR2F1 neurodevelopment and adult projection-neuron researchers could retest the pair with independent guides, DNA/protein editing assays, and NR2F1 rescue. Occupancy at Sorl1 and Sema7a, followed by neurite or synaptic phenotyping, would decide whether this is direct regulation with a functional consequence. Confirmation would motivate a compact adult-NR2F1 study.",128,144,3,3,2
bucket_3_rank_129,distributed_agents,B3,Distributed Agents catalog ranks 120-129,KLF7 perturbation links Tuba1a repression to a thalamic cytoskeletal transcript pattern,"In 151 TH Prkcd Grin2c Glut, Klf7 perturbation decreased Tuba1a, accompanied by directionally aligned changes in Coro1a, Sptan1, Stmn2, Map1b, Map6, and Dctn1. This pattern defines a cytoskeletal readout for studying KLF7-dependent projection maintenance in mature thalamic neurons. It also connects the adult perturbation response to KLF7's established developmental role in neuronal morphogenesis.",The result can nominate Tuba1a as a sensitive marker for testing KLF7 function in mature neurons. Its sparse support and lack of the expected Ntrk1/CDKN/Rac3 outputs make it incremental and low priority.,"KLF7 is expressed during neurogenesis (https://pubmed.ncbi.nlm.nih.gov/11336497/) and mouse genetics established roles in neuronal morphogenesis (https://pubmed.ncbi.nlm.nih.gov/15964824/). Those studies predicted axonal or cytoskeletal consequences, so the broad direction is not surprising even though the adult thalamic marker is new.","For Klf7-assigned nuclei from 151 TH Prkcd Grin2c Glut, Tuba1a decreased by log2FC -0.281 (FDR 0.0123). Coro1a (-0.432, FDR 0.0526), Sptan1 (-0.135, FDR 0.0902), Stmn2 (-0.248, FDR 0.180), Map1b (-0.193, FDR 0.464), Map6 (-0.220, FDR 0.690), and Dctn1 (-0.139, FDR 0.748) did not clear 5% FDR. Klf7 RNA itself was unchanged (-0.059, FDR 1.0), and Ntrk1, Cdkn1a, Cdkn1b, and Rac3 were unsupported; the apparent module could therefore reflect a single robust marker plus correlated nominal noise.","Axon-growth and KLF researchers could validate Tuba1a with independent Klf7 editing, KLF7 protein loss, and rescue, then test microtubule organization or axonal transport. Only if the weaker cytoskeletal genes and a cellular phenotype replicate would a larger mature-neuron KLF7 project be justified.",129,144,3,3,2
bucket_4_selection_01,codex,B4,Codex baseline findings,Dctn1 loss selectively activates a neuronal sterol-biosynthesis program,"The dynactin subunit perturbation Dctn1 upregulated SREBP/cholesterol genes rather than producing only a generic transport signature. The program was strongest in midbrain glutamatergic and pons/medulla GABA neurons and recurred across 12 neuronal classes, suggesting that dynein-dynactin disruption rapidly couples retrograde transport stress to lipid-homeostatic transcription in postmitotic neurons. This is a biological interpretation; the direct observation is concerted induction of Insig1, Msmo1, Hmgcs1, Fdps, Ldlr, Hmgcr, Nsdhl, Fdft1, Sc5d, and Sqle after Dctn1 perturbation.","DCTN1 is a causal gene for Perry syndrome and related ALS/frontotemporal/parkinsonian phenotypes, but the atlas points to sterol synthesis as an early neuronal response to partial loss of dynactin. If validated, this gives transport-neurodegeneration groups a concrete lipid axis to perturb genetically and pharmacologically before overt aggregate pathology appears.","DCTN1 was expected to affect axonal transport, stress granule dynamics, and TDP-43 pathology; DCTN1 disease reviews emphasize dynactin transport failure in Perry syndrome and motor-neuron disease (Konno et al., 2017, PMID: 28625595), and DCTN1 deficiency can exacerbate TDP-43 aggregation by impairing stress-granule resolution (Ueda et al., 2024, PMID: 38311779). A synchronized cholesterol-biosynthesis response in acute juvenile brain neurons was not the obvious expected output.","Dctn1 had 358 changed genes across 12 cell classes at FDR < 0.05 and |log2FC| >= 0.25. In 155 MB Glut, canonical sterol genes were among the top hits: Insig1 log2FC +2.07, q = 4.42e-55; Msmo1 +1.90, q = 1.38e-44; Hmgcs1 +1.61, q = 1.48e-41; Fdps +2.14, q = 3.29e-40; Ldlr +2.23, q = 1.34e-35; Hmgcr +1.48, q = 6.50e-34. The same module appeared in 191 MB P MY GABA, including Insig1 +1.80, q = 1.95e-27; Fdps +2.26, q = 6.82e-24; Msmo1 +1.86, q = 1.29e-22; and Dhcr24 +1.62, q = 1.65e-21. The result is transcriptionally coherent, but it is still a sparse-CRISPR RNA readout: the sterol increase may be compensatory, may reflect altered lysosomal cholesterol sensing, or may be a survivor bias among damaged cells.","ALS/FTD, Perry syndrome, axonal-transport, lipid-metabolism, and lysosome investigators would be the natural audience. The strongest follow-up program would combine acute Dctn1 knockout in defined midbrain neurons with SREBP reporter imaging, filipin or mass-spectrometry sterol measurements, dynein/dynactin rescue constructs, and tests of whether SREBP or cholesterol-synthesis inhibition modifies TDP-43 or alpha-synuclein stress.",1,30,5,4,5
bucket_4_selection_13,codex,B4,Codex baseline findings,Wdr26 loss produces a cerebellar GABA channel-and-splicing signature,"Wdr26 perturbation caused the largest cerebellar signature in the atlas, with 1,444 changed genes in 308 CB GABA. The observation is simultaneous loss of RNA-splicing and ion-channel transcripts such as Srrm2, Luc7l2, Kpna1, Cacna1g, Kalrn, Grik1, and Dagla, together with induction of Serpini1 and Kcnt1. The biological interpretation is that WDR26-containing CTLH/GID ubiquitin-ligase assemblies maintain postnatal cerebellar inhibitory-neuron excitability and RNA-processing state.","WDR26 haploinsufficiency causes Skraban-Deardorff neurodevelopmental syndrome, but the immediate vulnerable neuronal program has been unclear. This atlas suggests an adult-maintenance role in cerebellar GABA neurons, which shifts the disease model from general developmental delay toward a testable failure of channel and splicing homeostasis in inhibitory cerebellar circuits.","Recent structural work showed that Skraban-Deardorff mutations in WDR26 impair CTLH E3 ligase assembly and substrate-receptor functions (Gross et al., 2024, PMID: 38575527). That precedent predicted disrupted ubiquitin-ligase biology, not the specific Cacna1g/Grik1/Srrm2 program in cerebellar GABA neurons.","The Wdr26 signature was concentrated but large: 1,444 genes in 308 CB GABA, 470 in 155 MB Glut, and 324 in 191 MB P MY GABA. In 308 CB GABA, top effects included Srrm2 -0.799, q = 1.95e-11; Kalrn -1.28, q = 6.51e-10; Luc7l2 -0.716, q = 6.51e-10; Cacna1g -1.74, q = 6.51e-10; Kpna1 -0.784, q = 6.51e-10; Grik1 -1.46, q = 8.30e-09; and Kcnt1 +1.58, q = 7.48e-09. Srrm2 was also down in 155 MB Glut (-0.693, q = 7.09e-19) and 191 MB P MY GABA (-0.835, q = 4.68e-12). The caveat is that only five classes passed the signature threshold, so the cerebellar enrichment could reflect WDR26 biology, cerebellar sensitivity to CRISPR mosaicism, or both.","Skraban-Deardorff syndrome groups, cerebellar ataxia investigators, CTLH/GID complex biochemists, and channelopathy researchers would all have reason to pursue this. Promising grants would pair Wdr26 conditional deletion in cerebellar interneuron subclasses with patch-clamp tests of T-type calcium and BK sodium-activated potassium currents, rescue by wild-type versus syndrome WDR26 alleles, and proteomic identification of CTLH substrates controlling Srrm2 or channel expression.",2,30,5,4,5
bucket_4_selection_16,codex,B4,Codex baseline findings,Acute Tsc1 and Tsc2 loss couples mTOR derepression to a sterol-lysosome module in postnatal neurons,"Tsc1 and the lower-coverage Tsc2 perturbations converged on induction of lysosomal and cholesterol-associated transcripts, including Atp6v0b, Lamp2, Dpp7, Mcoln1, Cpeb1, Insig1, Cyp51, Msmo1, Ldlr, and Pik3r3. The observation appeared most strongly in 155 MB Glut and 151 TH Prkcd Grin2c Glut. The interpretation is that partial mosaic loss of the TSC complex is enough to engage an mTORC1-linked lysosomal/SREBP response in mature excitatory and inhibitory neurons.","TSC biology is central to epilepsy and autism, but this juvenile whole-brain Perturb-seq view tests the response outside embryonic cortical germline models. The data prioritize lipid/lysosome transcripts as rapid neuronal effectors to follow alongside cell-size and excitability phenotypes.","TSC1/TSC2 are established negative regulators of mTORC1; Tsc1 deletion reactivates mTOR and can promote adult CNS axon regeneration (Park et al., 2008, PMID: 18988856), and focal cortical models show that Tsc1, Pten, Depdc5, Rheb, and Mtor mutations have both convergent and gene-specific effects on cortical neurons (Nguyen et al., 2024, PMID: 38411613). A cholesterol/lysosome readout was plausible but not the canonical first phenotype.","Tsc1 produced 1,062 changed genes in 15 cell classes; Tsc2 produced 384 in 8. In 155 MB Glut, Tsc1 raised Cpeb1 +1.16, q = 2.44e-28; Atp6v0b +0.526, q = 1.40e-21; Pik3r3 +1.36, q = 2.26e-17; Lamp2 +0.782, q = 3.16e-12; and Insig1 +1.02, q = 2.22e-10. Tsc2 in the same cell class raised overlapping sterol and lysosome genes: Cyp51 +0.730, q = 6.42e-09; Cpeb1 +0.698, q = 1.39e-08; Insig1 +0.908, q = 7.05e-08; Atp6v0b +0.305, q = 9.52e-08; Msmo1 +0.873, q = 1.06e-07; and Ldlr +1.26, q = 3.05e-07. The interpretation is strengthened by pathway convergence but limited by transcript-only mTOR inference; phospho-S6, TFEB localization, and sterol abundance were not measured.","Tuberous sclerosis, mTORopathy, epilepsy, cortical malformation, and neuronal lipid biology groups would be highly motivated. A high-value program would perturb Tsc1 or Tsc2 in defined adult neuronal populations, jointly assay phospho-S6, SREBP2 processing, TFEB/TFE3 localization, lysosomal pH, and electrophysiology, then test whether lipid or lysosome modulators alter hyperexcitability independently of rapalogs.",3,30,4,4,5
bucket_4_selection_02,codex,B4,Codex baseline findings,Srsf1 loss elicits a pan-neuronal splice-factor autoregulatory response,"Srsf1 perturbation changed 8,969 genes and was active in all 23 cell classes. The direct observation is that Srsf1 mRNA and other splicing-associated transcripts such as Arglu1, Srsf2, Hnrnpdl, Dazap1, Rbm28, Gpatch8, Psip1, and Xpot were among the most recurrent responses. The interpretation is that postmitotic neurons rely on a tightly buffered SR-protein network, so weakening SRSF1 causes a broad compensatory and mis-splicing transcriptional state rather than a cell-type-limited phenotype.","This is the most pervasive RNA-processing signature in the atlas. It identifies SRSF1 as a central point of fragility in juvenile neurons and gives RNA-splicing researchers a cross-cell-type perturbational baseline for comparing disease RBPs such as TDP-43, MATR3, and CELF2.","SRSF1 is a core SR-family splicing regulator that negatively autoregulates its own productive splicing and buffers splicing-factor dosage (Ding et al., 2022, PMID: 35732114). Acute SRSF1 depletion can damage tissue homeostasis and produce broad transcriptional and proteomic consequences, including R-loop-associated stress in liver (Arif et al., 2023, PMID: 36759613). Broad effects were expected, but their uniformity across postnatal neuronal classes was not established.","At FDR < 0.05 and |log2FC| >= 0.25, Srsf1 had 2,390 changed genes in 151 TH Prkcd Grin2c Glut, 1,342 in 155 MB Glut, 1,092 in 191 MB P MY GABA, 826 in 217 P MY Pineal Glut, 720 in 066 CNU-HYa HY GABA, and smaller signatures in every remaining class. In 151 TH Prkcd Grin2c Glut, the strongest hits were Fam193b -1.48, q = 1.11e-80; Gpatch8 +1.02, q = 2.94e-73; Ptbp2 +1.19, q = 2.99e-72; Rbm28 +1.25, q = 4.09e-67; Arglu1 -1.04, q = 5.27e-65; and Srsf1 -1.83, q = 7.05e-65. Same-direction recurrence was high: Srsf1 and Arglu1 were down in 22 classes, Psip1 and Far1 were up in 21, and Fam193b, Trim17, and Rbm28 recurred in 20. The caveat is that DE of RNA regulators cannot reveal the causative isoform changes.","RNA-splicing groups, neuronal RBP labs, and teams building splice-modulating antisense or small-molecule therapies would be most interested. The next experiments are long-read isoform sequencing after Srsf1 deletion, rescue with splicing-active and inactive SRSF1, single-molecule RNA FISH for self-regulatory isoforms, and cell-class comparisons of the same perturbation in organoids and adult mouse brain.",4,30,3,4,5
bucket_4_selection_18,codex,B4,Codex baseline findings,Uba5 loss links neuronal UFMylation to mitochondrial translation and V-ATPase repression,"Perturbing the UFM1 E1 enzyme Uba5 changed 2,942 genes, concentrated in thalamic and midbrain/hindbrain neurons. The direct observation is induction of mitochondrial translation and oxidative-phosphorylation-associated RNAs such as Eef1a1, Ndufa2, Atp5j2, and Uqcc2, together with recurrent repression of Atp6v0c, Syt11, Srsf5, and Rsrp1. The interpretation is that loss of protein UFMylation forces neurons into an ER-mitochondrial compensation state coupled to reduced lysosomal acidification machinery.","UBA5 loss causes developmental and epileptic encephalopathy, but the postmitotic neuronal effector program is not well mapped. The atlas points to a discrete mitochondrial/V-ATPase module that can be followed by UFM1 conjugation proteomics and lysosome assays.","UFMylation downstream of UBA5 is known to be essential for development and ER stress resistance, and DEE44 variants in UBA5 reduce UFM1 activation to degrees that correlate with in vivo phenotypes in fly and biochemical assays (Pan et al., 2023, PMID: 38079206). The exact neuronal combination of mitochondrial induction plus Atp6v0c repression was not a standard prediction.","Uba5 had 2,177 changed genes in 151 TH Prkcd Grin2c Glut and 587 in 155 MB Glut. In thalamic neurons, strong effects included Eef1a1 +0.759, q = 4.76e-26; H1f4 +1.73, q = 4.76e-26; Rsrp1 -0.437, q = 7.65e-26; Abcc5 -0.500, q = 3.52e-25; Trim8 +1.05, q = 1.05e-23; and Ina +0.763, q = 4.00e-22. In 155 MB Glut, Atp6v0c was the top hit at -0.714, q = 1.11e-18, followed by Syt11 -0.580, q = 8.52e-18, Ndufa2 +0.945, q = 2.09e-11, Atp5j2 +0.806, q = 6.89e-10, and Uqcc2 +0.872, q = 1.68e-09. The UFMylation interpretation is coherent but still indirect because the atlas lacks UFM1-modified substrate measurements.","DEE44/epileptic encephalopathy groups, UFMylation biochemists, ER quality-control investigators, and mitochondrial neurologists are likely audiences. A strong research program would identify UFMylated ribosome-translocon substrates in neurons, then test whether lysosomal acidification or mitochondrial translation modifiers rescue firing and survival after Uba5 loss.",7,30,3,4,5
bucket_4_selection_14,codex,B4,Codex baseline findings,Thap11 perturbation reveals a conserved Rabac1-Sdccag8-Gdap1 mitochondrial/ciliary module,"Thap11 loss produced a distinctive signature in midbrain glutamatergic, thalamic glutamatergic, and subcortical GABA neurons. The same small gene set, including Rabac1, Sdccag8, Gdap1, Mdh1, Ndufa3, Acp6, Trub1, and Slc38a7, fell strongly, whereas Ttc41, Prkn, and Thoc6 rose. The biological interpretation is that THAP11 sustains a cilia/mitochondria/lysosome-adjacent maintenance module in mature neurons.","THAP11 is not as heavily studied in brain as CHD8, SETD1A, or KMT2D, so this repeated compact module is one of the more novel atlas signals. It nominates a tractable regulon connecting Sdccag8 ciliopathy biology, Gdap1 mitochondrial dynamics, and parkin-linked quality control.","Human neurodevelopmental disorders often converge on chromatin and autophagy regulation, and NDD-linked chromatin factors can regulate autophagy genes in non-neuronal contexts (Lewerissa et al., 2023, PMID:37674294). THAP11 itself had little direct precedent as an acute postmitotic neuronal regulator.","Thap11 perturbation changed 184 genes in 155 MB Glut with 327 perturbed nuclei, 373 genes in 151 TH Prkcd Grin2c Glut with 297 perturbed nuclei, 162 genes in 066 CNU-HYa HY GABA with 171 nuclei, and 125 genes in 191 MB P MY GABA with 136 nuclei. In 155 MB Glut, examples were Rabac1 -1.69, FDR 2.39e-62; Sdccag8 -1.90, FDR 7.82e-49; Gdap1 -1.96, FDR 2.07e-46; Ndufa3 -1.58, FDR 4.22e-42; Acp6 -2.13, FDR 5.42e-33; Ttc41 +3.98, FDR 2.16e-27; and Prkn +0.93, FDR 3.70e-27. The small repeated module is strong, but the target is a transcription factor and the evidence does not distinguish direct THAP11 binding from an indirect ciliary stress state.","Ciliopathy, mitochondrial-dynamics, and neuronal transcription-factor groups are the best fits. A high-value program would map THAP11 binding near Rabac1, Sdccag8, and Gdap1, image primary cilia and mitophagy in perturbed thalamic and midbrain neurons, and ask whether parkin induction is adaptive or a biomarker of mitochondrial injury.",9,30,5,4,4
bucket_4_selection_15,codex,B4,Codex baseline findings,NatB catalytic-subunit loss represses a Peg3-Kifc2-Snhg11 program in many neurons,"Naa20, the catalytic subunit of the NatB N-terminal acetyltransferase, produced 1,864 changed genes across 15 cell classes. The repeated observation is loss of Snhg11, Kifc2, Peg3, Repin1, Atp11b, Srsf5, Maged1, Cyp46a1, and Tafazzin, strongest in thalamic and midbrain neurons. The interpretation is that NatB-dependent co-translational acetylation stabilizes a neuronal RNA and membrane-trafficking maintenance program.","NatB acetylates a large fraction of proteins, and human NAA20 missense variants cause developmental delay and microcephaly, but neuron-resolved downstream genes are not known. This atlas identifies candidate NatB-sensitive RNAs that can be turned into substrate and rescue studies.","NAA20 and NAA25 form the NatB complex, whose N-terminal acetyltransferase activity requires complex assembly (Lasa et al., 2020, PMID: 32976911). Recessive NAA20 variants impair NatB formation/activity and cause developmental delay, intellectual disability, and microcephaly (Morrison et al., 2021, PMID: 34230638). The exact Peg3/Kifc2 neuronal module had no direct precedent.","The largest Naa20 signature was in 151 TH Prkcd Grin2c Glut, with 1,132 changed genes; recurrent same-direction hits included Snhg11 down in 9 classes, and Kifc2, Repin1, and Atp11b down in 8 each. In 151 TH Prkcd Grin2c Glut, top responses included Snhg11 -0.367, q = 1.14e-35; Peg3 -0.585, q = 2.53e-30; Gprasp1 -0.763, q = 1.54e-25; Aplp2 -0.573, q = 2.06e-25; Leng8 -0.475, q = 8.20e-25; Kifc2 -0.537, q = 3.20e-24; and Tafazzin -1.13, q = 7.59e-23. The core limitation is that none of these RNAs prove which NatB substrate is causative.","N-terminal acetylation, protein homeostasis, rare neurodevelopmental disease, and neuronal trafficking groups would be the main audience. Follow-up should combine N-terminomics after Naa20 CRISPR, substrate-specific acetylation rescue, and tests of whether KIFC2, TAFAZZIN, or cholesterol-turnover pathways mediate the transcriptional state.",10,30,5,4,4
bucket_4_selection_08,codex,B4,Codex baseline findings,Tpr loss induces a nuclear-transport transcriptional program centered on Ranbp6 and Xpo1,"The nuclear-pore basket gene Tpr produced 5,087 changed genes across 10 neuronal classes. The direct observation is induction of nuclear-transport and transcriptional regulators such as Ranbp6, Xpo1, Pigm, Tob1, Zfp770, Zfp281, Mga, and Med13, with repeated repression of Leng8, Amy1, Ogt, Rnpc3, and Wsb1. The interpretation is that partial nuclear-pore disruption is sensed as an RNA-export or nucleocytoplasmic-transport stress in postnatal neurons.","Nucleocytoplasmic transport has become a major theme in ALS and other neurodegenerative diseases. This atlas highlights TPR as a high-leverage perturbation in mammalian neurons and nominates compensatory transport factors, especially Ranbp6 and Xpo1, as measurable effectors.","Nuclear pore components and nucleocytoplasmic shuttling can modify TDP-43 localization and neurotoxicity in RNAi screens and neurons (Kim et al., 2012, PMID: 22872699). A direct juvenile-brain TPR perturbation signature, however, was not known.","Tpr changed 2,549 genes in 155 MB Glut, 1,672 in 151 TH Prkcd Grin2c Glut, 297 in 005 L4-5 IT CTX Glut, and 246 in 191 MB P MY GABA. In 155 MB Glut, Tob1 was +1.67, q = 1.55e-19; Chd9 +1.23, q = 5.23e-19; Ranbp6 +1.55, q = 3.27e-18; Leng8 -0.854, q = 1.35e-17; Amy1 -1.57, q = 9.71e-17; and Pigm +1.74, q = 5.76e-16. In 151 TH Prkcd Grin2c Glut, Xpo1 was +0.955, q = 3.20e-14 while Ogt was -0.626, q = 3.10e-13. The downstream labels fit transport stress, but the atlas does not measure RNA export or protein localization.","Nuclear-pore, RAN transport, ALS/FTD, repeat-expansion disease, and neuronal RNA-localization researchers would pursue this. Key experiments are single-molecule RNA export assays after Tpr loss, RAN gradient imaging, rescue with TPR basket-domain mutants, and genetic suppression screens for Ranbp6 or Xpo1.",11,30,4,4,4
bucket_4_selection_06,codex,B4,Codex baseline findings,Apc loss activates canonical Wnt feedback in Prkcd/Grin2c thalamic neurons,"Perturbing the beta-catenin destruction-complex component Apc in 151 TH Prkcd Grin2c Glut increased canonical Wnt feedback genes, including Axin2, Nkd1, and Notum, while altering neuronal transcripts such as Gls, Tagln3, Spock3, Bace2, Rgs16, and Olfm1. This is a clear positive control for Wnt derepression in a mature thalamic neuron class. The biological interpretation is that APC actively restrains beta-catenin target genes even weeks after neurogenesis.","Wnt signaling is usually framed around development, synapse formation, and plasticity. An acute postmitotic Apc signature in thalamus provides specific effectors for testing whether adult-like thalamic excitatory neurons keep Wnt feedback under active destruction-complex control.","APC is the canonical scaffold that promotes beta-catenin degradation, so Apc loss should increase Axin2 and Nkd1. The novelty is the in vivo cell class and the coupling of Wnt feedback to thalamic metabolic and synaptic transcripts.","Apc in 151 TH Prkcd Grin2c Glut used 187 perturbed nuclei, changed 263 genes, and had 71 stringent genes. The expected Wnt targets were prominent: Axin2 +1.25, FDR 2.43e-16; Nkd1 +2.45, FDR 2.14e-13; and Notum +1.56 among FDR-significant genes. Other high-scoring rows were Atrnl1 +0.64, FDR 9.18e-17; Serpine2 +1.56, FDR 1.94e-16; Gls -0.50, FDR 1.17e-12; and Bace2 +2.29, FDR 1.30e-10. The pathway direction is internally consistent, but the dataset cannot infer whether altered beta-catenin-dependent transcription is cell-autonomous at these loci without chromatin validation.","Wnt/beta-catenin neurobiology labs and thalamic development groups would be the main audience. Experiments should test beta-catenin stabilization by immunostaining, TCF/LEF reporter activation in Prkcd/Grin2c neurons, and whether Apc loss changes thalamic afferent or efferent synapse physiology.",12,30,4,4,4
bucket_4_selection_12,codex,B4,Codex baseline findings,Ogt loss represses RNA-processing and cholesterol-turnover transcripts while relieving Xpo1,"Perturbing O-GlcNAc transferase changed 1,472 genes across 19 classes. The direct observation is repeated loss of Ogt, Rsrp1, Snhg11, Kifc2, Odf2, Rbm14, Leng8, Oga, Cyp46a1, and Srrm2, together with induction of Xpo1 and Nap1l5. The interpretation is that O-GlcNAc cycling is required in postnatal neurons to maintain RNA-processing factors and cholesterol-turnover transcripts, not only cytosolic signaling proteins.",Neuronal O-GlcNAcylation has been linked to memory and brain aging; the atlas defines an early molecular state after acute OGT loss in many neuronal classes. The unexpected overlap with nuclear transport and RNA splicing makes Ogt a bridge between metabolic sensing and RNA homeostasis.,"Temporally controlled neuron-specific OGT knockout in mice recapitulates cellular and cognitive features of brain aging, whereas viral OGT overexpression can improve memory in aged hippocampus (Wheatley et al., 2019, PMID: 31588002). The prior did not specify a recurrent Rsrp1/Xpo1/Cyp46a1 signature.","The largest Ogt signatures were in 155 MB Glut (536 genes), 191 MB P MY GABA (316), and 066 CNU-HYa HY GABA (285). In 155 MB Glut, Ogt itself was -1.68, q = 3.09e-23; Xpo1 +1.52, q = 3.46e-20; Rsrp1 -0.901, q = 4.12e-17; Leng8 -0.898, q = 3.01e-15; Srrm2 -0.952, q = 1.89e-12; Odf2 -1.28, q = 1.89e-12; and Kifc2 -1.02, q = 7.71e-10. In 191 MB P MY GABA, Ogt was -2.42, q = 2.88e-21; Rsrp1 -1.05, q = 4.87e-11; Xpo1 +1.31, q = 5.73e-09; and Cyp46a1 -1.45, q = 6.08e-07. O-GlcNAcylation itself was not measured, so this is an inferred consequence of reduced OGT.","O-GlcNAc, neurodegeneration, brain aging, RNA-processing, and metabolic neuroscience groups are likely to follow this. The clean next step is to pair Ogt CRISPR with O-GlcNAc proteomics, OGA inhibition or substrate rescue, and isoform sequencing to test whether RSRP1/SRRM2 loss is transcriptional, splicing-mediated, or selection for damaged neurons.",12,30,4,4,4
bucket_4_selection_05,codex,B4,Codex baseline findings,Elp1 perturbation activates an amino-acid starvation and integrated-stress-like program in thalamic glutamatergic neurons,"Elp1 loss in 151 TH Prkcd Grin2c Glut up-regulated aminoacyl-tRNA synthetase, amino-acid transport, one-carbon, and ER-stress genes. The top induced genes included Cars, Asns, Shmt2, Slc1a4, Slc3a2, Slc38a2, Herpud1, Ddit3, and Mars1, while Adgrb1 was reduced. The result suggests that Elongator dysfunction in a mature thalamic class engages amino-acid deprivation and ER stress before overt cell loss.","ELP1 is central to familial dysautonomia and sensory/autonomic neuron development, but this postnatal central thalamic stress module is not the classical peripheral-neuron phenotype. It points to a convergent stress state that can be compared with tRNA-modification defects and integrated stress response inhibitors.","ELP1/IKBKAP is required cell-autonomously for sympathetic and sensory neuron innervation (Jackson et al., 2014, PMID:24917501), and mouse models show a requirement for Elp1 in visceral sensory peripheral and central circuitry (Tolman et al., 2022, PMID:35481599). A central postmitotic thalamic amino-acid response was plausible but not established.","Elp1 in 151 TH Prkcd Grin2c Glut used 153 perturbed nuclei, changed 570 genes, and had 152 genes at FDR < 1e-4. High-scoring rows were Cars +1.12, FDR 1.53e-28; Chac1 +3.43, FDR 2.28e-28; Asns +1.46, FDR 1.50e-27; Shmt2 +2.26, FDR 3.49e-22; Slc1a4 +2.23, FDR 2.87e-21; Herpud1 +0.98, FDR 4.76e-21; Adgrb1 -0.68, FDR 1.32e-20; Ddit3 +1.57, FDR 3.13e-20; and Slc38a2 +1.03, FDR 1.82e-16. The pattern is consistent with integrated stress, but ATF4 protein and eIF2alpha phosphorylation were not measured.","Familial dysautonomia investigators, Elongator/tRNA-modification groups, and integrated-stress-response labs would be interested. Follow-up should test ATF4/CHOP activation after Elp1 loss, profile tRNA wobble-uridine modifications in sorted neurons, and ask whether ISRIB-like modulation rescues the Asns/Chac1 module without worsening survival.",14,30,4,4,4
bucket_4_selection_10,codex,B4,Codex baseline findings,Kansl1 loss perturbs midbrain lysosomal and Golgi trafficking transcripts,"The NSL histone-acetyltransferase complex subunit Kansl1 produced a midbrain glutamatergic signature with decreased Nsmce2, Ccdc39, Smpd4, Ergic3, Stx8, Commd4, Exoc1, Pik3c3, Ctsa, and Man2b2, plus increases in genes such as Dgkz, Tspoap1, Son, and Ank2. The program points toward lysosome, ER-Golgi, and autophagy-associated traffic rather than a broad global transcriptional collapse. It implies that KANSL1 haploinsufficiency can acutely disturb membrane-turnover genes in postnatal midbrain neurons.","KANSL1 causes Koolen-de Vries syndrome and has been linked to autophagosome-lysosome fusion in mouse models. The atlas independently finds a lysosomal/traffic signature in a defined mature neuronal population, strengthening the case that autophagy-related rescue should be assessed in neurons and not only bulk tissue.","KANSL1 haploinsufficiency causes Koolen-de Vries syndrome; mouse Kansl1 haploinsufficiency impairs autophagosome-lysosome fusion and mitophagic clearance through transcriptional regulation of autophagy genes such as Stx17 (Li et al., 2022, PMID:35177641). Thus a trafficking signature was expected, but its exact midbrain form was new.","Kansl1 in 155 MB Glut used 113 perturbed nuclei, changed 350 genes, and had 39 stringent genes. Examples included Nsmce2 -1.33, FDR 2.62e-10; Ccdc39 -1.81, FDR 1.04e-09; Ttc4 -0.80, FDR 8.55e-09; Dgkz +0.82, FDR 1.02e-08; Smpd4 -1.13, FDR 1.10e-08; Son +0.54, FDR 2.33e-07; Ergic3 -0.89, FDR 2.79e-07; Stx8 -1.19, FDR 3.41e-07; and Commd4 -1.15, FDR 6.34e-07. The signature is weaker than Setd1a or Sin3a and lacks direct lysosomal flux measurements.","Koolen-de Vries syndrome, autophagy, and chromatin-regulatory-complex groups are the likely audience. Follow-ups should test whether retinoic-acid or lysosomal-fusion interventions rescue the midbrain signature, and should map KANSL1/KAT8 histone-acetylation changes near vesicle-fusion genes.",16,30,4,4,4
bucket_4_selection_20,codex,B4,Codex baseline findings,Atp6v0c loss coordinately reduces V-ATPase and calcium-channel transcripts,"The V-ATPase V0 subunit perturbation Atp6v0c had an unusually recurrent target response, with Atp6v0c itself lower in 22 classes and Atp6v0a1 lower in 12. The strongest downstream effects were in 151 TH Prkcd Grin2c Glut, where calcium and GABA-related transcripts including Atp2a2, Prkcg, Prkcd, and Gabrd were also repressed. The interpretation is that partial lysosomal proton-pump disruption feeds back on V-ATPase subunit expression and neuronal excitability genes.",V-ATPase genes are central to lysosome biology but have been hard to place in specific neuronal transcriptomes. This perturbation nominates a cell-class-specific acidification-to-excitability path that can be validated with lysosomal pH and channel physiology.,"The TFEB/V-ATPase axis is known to coordinate lysosomal function and development, and ATP6V0C impairment can alter the endolysosomal compartment in human epithelial cells (Tognon et al., 2016, PMID: 26727288). A coherent thalamic Prkcd/Gabrd calcium-channel module was less expected.","Atp6v0c produced 755 changed genes across 22 classes, with 490 in 151 TH Prkcd Grin2c Glut and 151 in 052 Pvalb Gaba. In thalamic neurons, Atp6v0c was -1.43, q = 1.04e-34; Atp2a2 -0.577, q = 3.09e-20; Atp6v0a1 -0.577, q = 2.39e-19; Pcp4 -0.957, q = 1.36e-17; Prkcg -0.856, q = 1.56e-16; Cadm4 -0.758, q = 1.14e-13; Prkcd -0.699, q = 1.22e-12; and Gabrd -0.697, q = 1.71e-12. Atp6v0a1 was also down in Pvalb neurons (-0.450, q = 1.29e-07). Because V-ATPase loss can alter viability, late degenerative repression is a plausible alternative.","Lysosome, V-ATPase, epileptic encephalopathy, and thalamic-circuit groups would pursue this. Follow-up should quantify lysosomal pH and membrane potential after Atp6v0c loss, rescue with pump-dead versus pump-active ATP6V0C, and measure whether GABA-A delta subunit and PRKCG loss change thalamic firing.",18,30,3,4,4
bucket_4_selection_11,codex,B4,Codex baseline findings,Rab7 loss elicits a lysosomal hydrolase and sterol-response state,"Rab7 perturbation was dominated by a thalamic signature that induced Atp6v1b2, Dpp7, Scpep1, Insig1, Atp6v0b, Slc20a1, and Dync1h1, with a smaller matching midbrain program containing Lamp2, Grn, Mcoln1, and Rab7 itself. The observation supports a model in which late-endosome transport stress triggers lysosomal and cholesterol-sensing compensation in neurons. The interpretation is narrower than for TSC: Rab7 likely affects endosome-lysosome maturation upstream of the sterol module.","The Rab7 response connects a core trafficking GTPase to the same lysosome/sterol genes seen after TSC perturbation, providing an epistasis opportunity. If RAB7 and mTORC1 converge on Atp6v0b, Dpp7, and Insig1, this could identify druggable lysosomal feedback that is common to several disease genes.","RAB7 is required for endosome-lysosome trafficking and lysosome reformation; in GBA1-deficient Parkinson disease models, Rab7 accumulates with lysosomal recycling failure and altered cholesterol homeostasis (Magalhaes et al., 2016, PMID: 27378698). The exact acute neuron signature was not directly known.","Rab7 had 436 changed genes across 7 cell classes, with 361 in 151 TH Prkcd Grin2c Glut. In that thalamic class, Atp6v1b2 was +0.443, q = 1.06e-16; Dpp7 +0.904, q = 6.11e-16; Scpep1 +1.57, q = 5.18e-15; Dusp1 +1.25, q = 1.20e-14; Insig1 +0.948, q = 1.87e-13; and Atp6v0b +0.322, q = 4.26e-13. In 155 MB Glut, Atp6v0b was +0.372, q = 4.15e-11; Dpp7 +0.888, q = 7.29e-08; Lamp2 +0.600, q = 1.44e-07; Grn +0.935, q = 2.18e-07; and Mcoln1 +0.455, q = 1.67e-06. Support is lower than for TSC because one thalamic class accounts for most DE genes.","Endolysosomal trafficking, Parkinson disease, autophagy, and neuronal lipid investigators would be the main audience. The most informative follow-up would test Rab7 x Tsc1 genetic interactions, live-cell lysosome motility, mTORC1 recruitment to lysosomes, and sterol reporters in the same neuronal subtypes.",19,30,4,3,4
bucket_4_selection_09,codex,B4,Codex baseline findings,"Rnpc3 loss creates a core spliceosome response centered on Mapk8, Ap2a1, Wdfy1, Pcyt2, and Pten","Rnpc3, a U11/U12 spliceosome component, produced its strongest response in 151 TH Prkcd Grin2c Glut, with additional midbrain and GABAergic responses. The thalamic signature decreased Rnpc3 and increased Mapk8, Ap2a1, Pcyt2, Wdfy1, Srp72, Trit1, Vash2, and Pten. The interpretation is that minor-spliceosome stress in mature thalamic neurons engages JNK, clathrin-adaptor, phospholipid, and PTEN-associated pathways.","This finding broadens the RNA-processing theme beyond SR factors and identifies a distinct downstream pattern from Srsf1. Because minor introns are enriched in regulatory genes, this is a useful hypothesis for why partial defects in a rare spliceosome component might have amplified neuronal consequences.","Alternative splicing is central to neuronal maturation, and SRSF1/PTBP2-like cascades were already known in developing cortex (Weber et al., 2023, PMID:37697438). A U11/U12-specific factor such as RNPC3 would be expected to cause RNA-processing stress, but the Mapk8/Ap2a1/Pten axis is more specific.","Rnpc3 in 151 TH Prkcd Grin2c Glut used 149 perturbed nuclei, changed 820 genes, and retained 236 stringent genes. Top rows were Rnpc3 -1.69, FDR 3.22e-39; Amy1 -2.10, FDR 1.84e-37; Mapk8 +1.76, FDR 6.03e-37; Ap2a1 +1.37, FDR 3.93e-36; Pcyt2 +1.82, FDR 1.30e-34; Wdfy1 +1.89, FDR 2.71e-33; and Pten +0.92, FDR 2.47e-21. Rnpc3 also changed 221 genes in 155 MB Glut, 77 in 052 Pvalb Gaba, 73 in 066 CNU-HYa HY GABA, and 55 in 191 MB P MY GABA. Long-read sequencing is required to show that minor intron retention drives these abundance changes.","Minor-spliceosome, RNA surveillance, and neuronal PTEN/mTOR researchers would be motivated. Experiments should quantify U12 intron retention, test JNK inhibition, and ask whether Pten induction is a compensatory brake on stress-induced growth pathways.",21,30,3,4,4
bucket_4_selection_07,codex,B4,Codex baseline findings,U2af2 perturbation depresses spliceosomal and calcium-handling genes in thalamic and midbrain neurons,"The major-spliceosome factor U2af2 produced parallel responses in thalamic, midbrain glutamatergic, and midbrain GABA classes. In 151 TH Prkcd Grin2c Glut, the strongest decreased genes included Rsrp1, Srrm2, Atp2a2, Agrn, Luc7l2, Srsf11, and Atp11b, while Map7d1, Iffo1, and Copb1 increased. This supports a second, mechanistically distinct spliceosome-to-synaptic-homeostasis relationship.","U2af2 sits at the 3-prime splice-site recognition step, so its in vivo downstream genes can separate general spliceosome failure from SRSF1-specific or minor-spliceosome programs. The thalamic and midbrain signatures nominate Atp2a2 and Agrn as vulnerable mature-neuron transcripts after splice-site recognition stress.","Splicing factors have broad RNA targets and are essential for neuronal isoform diversity; published transcriptome compendia have shown that perturbing splicing factors creates identifiable expression signatures (Yu et al., 2020, PMID:32546682). A large signature after U2af2 loss is not surprising, but the recurrent calcium and extracellular-matrix components add specificity.","U2af2 changed 650 genes in 151 TH Prkcd Grin2c Glut with 222 perturbed nuclei, 744 in 155 MB Glut with 187 nuclei, and 398 in 191 MB P MY GABA with 85 nuclei. Thalamic high-scoring rows included Rsrp1 -0.98, FDR 9.01e-29; Srrm2 -0.44, FDR 8.73e-17; Map7d1 +0.65, FDR 8.73e-17; Atp2a2 -0.47, FDR 9.74e-16; Agrn -0.56, FDR 1.13e-14; Luc7l2 -0.45, FDR 9.39e-13; and Atp11b -0.72, FDR 1.01e-12. As with Srsf1, the snRNA-seq assay cannot distinguish isoform switching from total mRNA changes.","Spliceosome structural groups, RNA-therapeutic developers, and neuronal calcium-homeostasis labs would be interested. Follow-up should use nanopore cDNA or targeted RT-PCR panels to test whether Atp2a2 or Agrn isoforms are direct U2AF2-sensitive transcripts.",22,30,3,4,4
bucket_4_selection_04,codex,B4,Codex baseline findings,"Prdm10 perturbation reveals a conserved Madd, Cadm3, Ptov1, Stmn2, Myt1, and Uchl1 repression signature","Prdm10 disruption generated a small but very coherent signature across 10 classes. The observed result was recurrent down-regulation of Madd, Cadm3, Ptov1, Stmn2, Myt1, Uchl1, Them6, and Ackr1, with Madd significant in all 10 classes and Cadm3 and Ptov1 in 9. The biological interpretation is that PRDM10 sustains a mature axonal and cell-adhesion gene module in multiple neuronal types.",PRDM-family chromatin factors are plausible disease genes but PRDM10 has much less mature-neuron biology than many regulators in the screen. The atlas elevates it by identifying a precise downstream module that recurs despite only 182 significant rows.,"PRDM proteins often act as transcriptional and chromatin regulators, so a targeted gene-expression effect was plausible. PubMed did not reveal a direct neuronal Prdm10 -> Madd/Cadm3/Ptov1 precedent, making the relationship a relatively novel atlas-derived finding.","Prdm10 had significant signatures in 151 TH Prkcd Grin2c Glut, 191 MB P MY GABA, 155 MB Glut, 066 CNU-HYa HY GABA, 145 MH-LH TH Glut, 110 CNU-HYa HY MM Glut, and 052 Pvalb Gaba. Madd was down in 10 classes, median log2FC -1.28, min q=8.7e-42; Cadm3 was down in 9, median -0.99, min q=2.3e-36; Ptov1 was down in 9, median -1.27, min q=3.6e-30; and Stmn2 was down in 7, min q=7.4e-18. The signature is compact, so it needs more sensitivity to reveal whether PRDM10 has broader direct targets.","Chromatin and axonal maintenance groups would be best suited to pursue this. A strong experiment would validate PRDM10 binding at Madd, Cadm3, and Stmn2, then assay axon maintenance and cell-adhesion phenotypes after direct disruption of each candidate effector.",24,30,4,4,4
bucket_4_selection_19,codex,B4,Codex baseline findings,"Rora loss acutely represses Prkcg, Rasa4, Tnnt1, and Pcp4 in Grin2c thalamic neurons","The orphan nuclear receptor perturbation Rora yielded 261 changed genes, all in 151 TH Prkcd Grin2c Glut. The observation is downregulation of Prkcg, Adgrb1, Rasa4, Tnnt1, Plekhd1, Elmod1, Atp6ap2, Pcp4, and Kcnc2. The interpretation is that RORA, although classically tied to Purkinje cells, has an acute maintenance role in a non-cerebellar thalamic excitability program.","This is a lower-breadth but provocative extension of RORA biology outside cerebellar Purkinje neurons. If validated, it argues that RORA-targeting ataxia therapeutics may need to account for thalamic neurons and their calcium/kinase transcripts.","RORalpha is required for Purkinje-cell development and survival, and reduced RORalpha worsens SCA1 pathogenesis (Serra et al., 2006, PMID: 17110330; Doulazmi et al., 2006, PMID: 16807193). Purkinje vulnerability was expected; an acute Grin2c thalamic signature was not.","The only active class was 151 TH Prkcd Grin2c Glut, where Prkcg was -0.764, q = 9.68e-18; Adgrb1 -0.702, q = 1.35e-17; Rasa4 -1.54, q = 3.98e-16; Tnnt1 -1.70, q = 2.53e-14; Plekhd1 -1.51, q = 1.44e-13; Elmod1 -1.14, q = 6.61e-13; Atp6ap2 -0.643, q = 7.11e-13; Pcp4 -0.584, q = 3.72e-11; and Kcnc2 -0.355, q = 3.72e-10. The support score is only 3 because the effect is confined to one atlas class and Rora itself did not pass the significance threshold in this filtered table.","RORalpha/SCA, thalamic excitability, nuclear-receptor pharmacology, and calcium-signaling groups are the likely audience. The most direct validation would use RNAscope and RORA ChIP/CUT&Tag in Prkcd/Grin2c thalamus, then test whether RORalpha agonists rescue the Prkcg/Pcp4/Kcnc2 state.",24,30,4,3,4
bucket_4_selection_03,codex,B4,Codex baseline findings,Zfp423 loss disrupts Grid2ip/Cd47 and derepresses Ntsr1 in Prkcd/Grin2c thalamic neurons,"Zfp423 perturbation in 151 TH Prkcd Grin2c Glut reduced Grid2ip, Cd47, Syt13, Cnnm1, Pxdn, Mpi, Adgrb2, and Grin2c, and strongly increased Ntsr1, Kcnab3, Nptxr, and Scn1b. The observation suggests that ZFP423, known for developmental cerebellar functions, continues to regulate a postnatal thalamic identity module and suppresses corticothalamic-like Ntsr1 expression in this thalamic class.",ZFP423 is classically tied to cerebellar development and ciliopathy/Joubert biology. A mature Prkcd/Grin2c thalamic response is therefore a high-novelty lead that could reveal an unappreciated role for this zinc-finger factor outside cerebellar progenitors and Purkinje cells.,"Zfp423 is required for normal cerebellar development in mice (Warming et al., 2006, PMID:16943432), and ZFP423/ZNF423 regulates cell-cycle progression and DNA-damage responses in Purkinje neuron progenitors (Casoni et al., 2017, PMID:28893945). Those studies did not predict a strong postnatal thalamic Grid2ip/Ntsr1 relationship.","Zfp423 in 151 TH Prkcd Grin2c Glut used 160 perturbed nuclei, changed 139 genes, and retained 40 stringent genes. Top rows were Grid2ip -1.27, FDR 3.07e-45; Cd47 -0.99, FDR 1.27e-35; Ntsr1 +5.14, FDR 5.89e-21; Kcnab3 +1.39, FDR 4.34e-17; Syt13 -0.89, FDR 2.62e-15; Nptxr +1.00, FDR 4.98e-15; Pxdn -1.47, FDR 8.79e-14; Scn1b +0.37, FDR 1.09e-10; and Grin2c -0.57, FDR 1.65e-10. The signature is specific but smaller than top-ranked chromatin signatures, and the increase in a sparse marker such as Ntsr1 needs single-molecule validation.","Cerebellar-development, ciliopathy, and thalamic-identity groups would be the best audience. Experiments should map ZFP423 binding near Grid2ip and Ntsr1, repeat the perturbation in adult thalamus, and ask whether ZFP423 loss shifts electrophysiological identity toward a corticothalamic-like state.",25,30,5,3,4
bucket_4_selection_17,codex,B4,Codex baseline findings,"TDP-43 loss induces Ppp6c, Gabpa, and trafficking transcripts while repressing Snrnp70","Tardbp perturbation produced 533 changed genes across 16 classes. The direct observation is downregulation of Tardbp and Snrnp70, with upregulation of Ppp6c, Fam219a, Brms1l, Uggt2, Usp15, Gabpa, Abcc5, Fam135a, Sec24b, Kifbp, and Unc13a. The interpretation is that partial nuclear TDP-43 loss causes a compensatory phosphatase/trafficking RNA program in multiple postnatal neuronal classes.","TDP-43 nuclear depletion is central to ALS/FTD. Although this atlas does not assay cryptic exons, it offers an in vivo mouse brain perturbation signature that can be aligned against human TDP-43 loss-of-function datasets.","Nuclear TDP-43 represses cryptic exon inclusion, with direct disease links through UNC13A and STMN2 misprocessing (Ma et al., 2022, PMID: 35197626; Baughn et al., 2023, PMID: 36927019). A coding-gene DE signature was plausible but less established than splicing changes.","Tardbp changed 187 genes in 151 TH Prkcd Grin2c Glut, 96 in 191 MB P MY GABA, and 89 in 155 MB Glut, with smaller signatures across 13 more classes. In thalamic neurons, Ppp6c was +1.98, q = 2.45e-41; Tardbp -1.41, q = 1.60e-39; Gabpa +2.25, q = 1.74e-38; Usp15 +0.978, q = 8.87e-37; Fam135a +2.13, q = 4.16e-35; Brms1l +1.24, q = 2.02e-32; Uggt2 +1.36, q = 7.38e-32; and Abcc5 +0.548, q = 3.03e-27. Ppp6c and Fam135a repeated in 191 MB P MY GABA. The support is incomplete because the canonical cryptic-splicing consequences of TDP-43 loss are invisible in this DE table.","ALS/FTD, TDP-43, neuronal RNA-splicing, and antisense-therapy groups are obvious. Follow-up should perform long-read and targeted cryptic-exon assays on the same Tardbp-perturbed nuclei, then compare whether the DE signature tracks with UNC13A/STMN2-like splicing rescue or with separate cellular stress.",25,30,3,3,4