generated: '2026-08-23'
method: searched
source: >-
  Searched npm (registry.npmjs.org search + registry metadata), PyPI (pypi.org/pypi/<name>/json),
  and the company's public GitHub organization https://github.com/ksqtx via the GitHub API on
  2026-08-23.
note: >-
  KSQ Therapeutics publishes no API client library, because it publishes no API. It does publish
  first-party open-source bioinformatics code in its own GitHub organization, and that is what is
  recorded here. Nothing below is an SDK, so NO SDKs pointer is emitted in apis.yml — wiring one
  would fire the sdk_count ergonomics checks for client libraries that do not exist.
sdk_count: 0
registry_search:
- registry: npm
  query: ksq therapeutics
  result: no first-party package
  status: 200
- registry: pypi
  probed: https://pypi.org/pypi/ksq/json
  status: 404
- registry: pypi
  probed: https://pypi.org/pypi/trace/json
  status: 404
  note: The name the TRACE repo declares in pyproject.toml is unclaimed on PyPI.
- registry: pypi
  probed: https://pypi.org/pypi/tracepy/json
  status: 200
  note: >-
    NOT KSQ. pypi.org/project/tracepy is "Optical design software for python" by Gavin Niendorf
    (github.com/GNiendorf/tracepy), last released 2025-02-03. It shares the import package name
    declared in the KSQ repo and is unrelated. Recorded here so a future pass does not misattribute
    it.
packages:
- language: python
  registry: source
  name: trace
  import_package: tracepy
  url: https://github.com/ksqtx/TRACE
  install: pip install git+https://github.com/ksqtx/TRACE.git
  version: 0.1.0
  published: null
  official: true
  license: GPL-3.0 (repository) / MIT (declared in pyproject.toml — the two disagree)
  description: >-
    TRACE — Tumor Reactivity Assessment using Clonal Expression. A machine-learning system for
    single-cell RNA-seq classification, built on scanpy/anndata, scikit-learn, XGBoost, Optuna,
    SHAP and MLflow, with a 13-command Click CLI. Directly adjacent to KSQ's eTIL programs.
  note: >-
    version 0.1.0 read from pyproject.toml at github.com/ksqtx/TRACE@main on 2026-08-23. published
    is null because this package has never been released to any registry — there is no metadata
    endpoint to query and no tagged release on the repository, so a consumer can only install from a
    git ref. Last commit pushed 2026-04-14. Note the repository's [project.urls] block still points
    at github.com/trace/trace and its author metadata reads "TRACE Team <team@trace.dev>" — unfilled
    template placeholders, not KSQ contact details. Ownership rests on the repo living in the ksqtx
    organization (github.com/ksqtx, org display name "KSQ Therapeutics") and on the subject matter
    matching KSQ's published eTIL work; it is NOT asserted from the pyproject metadata.
- language: r
  registry: source
  name: HRD-predictions-cell-lines
  url: https://github.com/ksqtx/HRD-predictions-cell-lines
  install: git clone https://github.com/ksqtx/HRD-predictions-cell-lines.git
  version: null
  published: null
  official: true
  license: GPL-3.0
  description: >-
    Data and R code accompanying "Pan-cancer Analysis of Homologous Recombination Deficiency in Cell
    Lines" — CHORD predictions and HRDsum scores.
  note: >-
    version null: this is a paper companion repository, not a packaged library. It has no
    DESCRIPTION, no version field and no registry release, so there is nothing to read and nothing to
    guess. Last pushed 2024-06-26.
forks_not_packages:
- gatk
- nf-core-rnaseq
- airrflow
- farver
- shinyTree
- ksqShinyTree
- awesome-microsoft-copilot-prompts
forks_note: >-
  The remaining seven repositories in github.com/ksqtx are forks of third-party projects (Broad
  Institute GATK, nf-core pipelines, R utility packages, a prompt library). They are not KSQ
  software and are not recorded as packages.