{
"cells": [
{
"cell_type": "markdown",
"metadata": {},
"source": [
"# Palantir Basic Tutorial"
]
},
{
"cell_type": "markdown",
"metadata": {},
"source": [
"### Table of contents\n",
" - [Introduction](##Introduction)\n",
" - [Loading data](##Loading-data)\n",
" - [Data Processing](##Data-Processing)\n",
" - [Running Palantir](##Running-Palantir)\n",
" - [Visualizing Palantir results](##Introduction)\n",
" - [Gene expression trends](##Introduction)\n",
" - [Clustering of gene expression trends](##Introduction)"
]
},
{
"cell_type": "markdown",
"metadata": {},
"source": [
"## Introduction"
]
},
{
"cell_type": "markdown",
"metadata": {},
"source": [
"Palantir is an algorithm to align cells along differentiation trajectories. Palantir models differentiation as a stochastic process where stem cells differentiate to terminally differentiated cells by a series of steps through a low dimensional phenotypic manifold. Palantir effectively captures the continuity in cell states and the stochasticity in cell fate determination. \n",
"\n",
"See our manuscript for more details."
]
},
{
"cell_type": "markdown",
"metadata": {
"tags": []
},
"source": [
"## Imports"
]
},
{
"cell_type": "code",
"execution_count": 1,
"metadata": {
"ExecuteTime": {
"end_time": "2022-06-15T17:45:34.735085Z",
"start_time": "2022-06-15T17:45:32.748529Z"
},
"execution": {
"iopub.execute_input": "2023-11-07T00:43:36.749311Z",
"iopub.status.busy": "2023-11-07T00:43:36.748777Z",
"iopub.status.idle": "2023-11-07T00:43:44.437350Z",
"shell.execute_reply": "2023-11-07T00:43:44.436365Z",
"shell.execute_reply.started": "2023-11-07T00:43:36.749278Z"
},
"tags": []
},
"outputs": [],
"source": [
"import palantir\n",
"import scanpy as sc\n",
"import pandas as pd\n",
"import os\n",
"\n",
"# Plotting\n",
"import matplotlib\n",
"import matplotlib.pyplot as plt\n",
"\n",
"# warnings\n",
"import warnings\n",
"from numba.core.errors import NumbaDeprecationWarning\n",
"\n",
"warnings.filterwarnings(action=\"ignore\", category=NumbaDeprecationWarning)\n",
"warnings.filterwarnings(\n",
" action=\"ignore\", module=\"scanpy\", message=\"No data for colormapping\"\n",
")\n",
"\n",
"# Inline plotting\n",
"%matplotlib inline"
]
},
{
"cell_type": "markdown",
"metadata": {},
"source": [
"## Loading data"
]
},
{
"cell_type": "markdown",
"metadata": {},
"source": [
"We recommend the use of scanpy Anndata objects as the preferred mode of loading and filtering data. "
]
},
{
"cell_type": "markdown",
"metadata": {},
"source": [
"A sample RNA-seq csv data is available at . This sample data will be used to demonstrate the utilization and capabilities of the Palantir package. This dataset contains ~4k cells and ~16k genes and is pre-filtered. Check the scanpy introductory tutorial for filtering cells and genes. \n",
"\n"
]
},
{
"cell_type": "code",
"execution_count": 2,
"metadata": {
"ExecuteTime": {
"end_time": "2022-06-15T17:45:34.854759Z",
"start_time": "2022-06-15T17:45:34.737139Z"
},
"execution": {
"iopub.execute_input": "2023-11-07T00:43:44.440337Z",
"iopub.status.busy": "2023-11-07T00:43:44.439352Z",
"iopub.status.idle": "2023-11-07T00:43:44.601584Z",
"shell.execute_reply": "2023-11-07T00:43:44.600852Z",
"shell.execute_reply.started": "2023-11-07T00:43:44.440302Z"
},
"tags": []
},
"outputs": [
{
"data": {
"text/plain": [
"AnnData object with n_obs × n_vars = 4142 × 16106"
]
},
"execution_count": 2,
"metadata": {},
"output_type": "execute_result"
}
],
"source": [
"# Load sample data\n",
"data_dir = os.path.expanduser(\"./\")\n",
"download_url = \"https://dp-lab-data-public.s3.amazonaws.com/palantir/marrow_sample_scseq_counts.h5ad\"\n",
"file_path = os.path.join(data_dir, \"marrow_sample_scseq_counts.h5ad\")\n",
"ad = sc.read(file_path, backup_url=download_url)\n",
"ad"
]
},
{
"cell_type": "markdown",
"metadata": {},
"source": [
"NOTE: Counts are assumed to the normalized. If you have already normalized the data, skip past the `Normalization` section"
]
},
{
"cell_type": "markdown",
"metadata": {},
"source": [
"## Data processing"
]
},
{
"cell_type": "markdown",
"metadata": {},
"source": [
"### Normalization"
]
},
{
"cell_type": "markdown",
"metadata": {},
"source": [
"Normalize the data for molecule count distribution using the `scanpy` interface"
]
},
{
"cell_type": "code",
"execution_count": 3,
"metadata": {
"ExecuteTime": {
"end_time": "2022-06-15T17:45:35.018205Z",
"start_time": "2022-06-15T17:45:34.856894Z"
},
"execution": {
"iopub.execute_input": "2023-11-07T00:43:44.602873Z",
"iopub.status.busy": "2023-11-07T00:43:44.602528Z",
"iopub.status.idle": "2023-11-07T00:43:44.728116Z",
"shell.execute_reply": "2023-11-07T00:43:44.727055Z",
"shell.execute_reply.started": "2023-11-07T00:43:44.602851Z"
},
"tags": []
},
"outputs": [],
"source": [
"sc.pp.normalize_per_cell(ad)"
]
},
{
"cell_type": "markdown",
"metadata": {},
"source": [
"We recommend that the data be log transformed. Note that, some datasets show better signal in the linear scale while others show stronger signal in the log scale.\n",
"\n",
"The function below uses a `pseudocount` of 0.1 instead of 1."
]
},
{
"cell_type": "code",
"execution_count": 4,
"metadata": {
"ExecuteTime": {
"end_time": "2022-06-15T17:45:35.098276Z",
"start_time": "2022-06-15T17:45:35.020719Z"
},
"execution": {
"iopub.execute_input": "2023-11-07T00:43:44.729555Z",
"iopub.status.busy": "2023-11-07T00:43:44.729153Z",
"iopub.status.idle": "2023-11-07T00:43:44.761214Z",
"shell.execute_reply": "2023-11-07T00:43:44.760235Z",
"shell.execute_reply.started": "2023-11-07T00:43:44.729533Z"
},
"tags": []
},
"outputs": [],
"source": [
"palantir.preprocess.log_transform(ad)"
]
},
{
"cell_type": "markdown",
"metadata": {},
"source": [
"### Highly variable gene selection"
]
},
{
"cell_type": "markdown",
"metadata": {},
"source": [
"Highly variable gene selection can also be performed using the `scanpy` interface"
]
},
{
"cell_type": "code",
"execution_count": 5,
"metadata": {
"ExecuteTime": {
"end_time": "2022-06-15T17:45:36.297603Z",
"start_time": "2022-06-15T17:45:35.614883Z"
},
"execution": {
"iopub.execute_input": "2023-11-07T00:43:44.762283Z",
"iopub.status.busy": "2023-11-07T00:43:44.762070Z",
"iopub.status.idle": "2023-11-07T00:43:45.138555Z",
"shell.execute_reply": "2023-11-07T00:43:45.137821Z",
"shell.execute_reply.started": "2023-11-07T00:43:44.762265Z"
},
"tags": []
},
"outputs": [],
"source": [
"sc.pp.highly_variable_genes(ad, n_top_genes=1500, flavor=\"cell_ranger\")"
]
},
{
"cell_type": "markdown",
"metadata": {},
"source": [
"### PCA"
]
},
{
"cell_type": "markdown",
"metadata": {},
"source": [
"PCA is the first step in data processing for Palantir. This representation is necessary to overcome the extensive dropouts that are pervasive in single cell RNA-seq data. \n",
"\n",
"Rather than use a fixed number of PCs, we recommend the use of components that explain 85% of the variance in the data after highly variable gene selection."
]
},
{
"cell_type": "code",
"execution_count": 6,
"metadata": {
"ExecuteTime": {
"end_time": "2022-06-15T17:45:36.694677Z",
"start_time": "2022-06-15T17:45:36.299597Z"
},
"execution": {
"iopub.execute_input": "2023-11-07T00:43:45.139455Z",
"iopub.status.busy": "2023-11-07T00:43:45.139281Z",
"iopub.status.idle": "2023-11-07T00:43:45.461286Z",
"shell.execute_reply": "2023-11-07T00:43:45.460319Z",
"shell.execute_reply.started": "2023-11-07T00:43:45.139441Z"
},
"tags": []
},
"outputs": [],
"source": [
"# Note in the manuscript, we did not use highly variable genes but scanpy by default uses only highly variable genes\n",
"sc.pp.pca(ad)"
]
},
{
"cell_type": "code",
"execution_count": 7,
"metadata": {
"ExecuteTime": {
"end_time": "2022-06-15T17:45:36.701107Z",
"start_time": "2022-06-15T17:45:36.697194Z"
},
"execution": {
"iopub.execute_input": "2023-11-07T00:43:45.463312Z",
"iopub.status.busy": "2023-11-07T00:43:45.463001Z",
"iopub.status.idle": "2023-11-07T00:43:45.468680Z",
"shell.execute_reply": "2023-11-07T00:43:45.467554Z",
"shell.execute_reply.started": "2023-11-07T00:43:45.463295Z"
},
"tags": []
},
"outputs": [
{
"data": {
"text/plain": [
"AnnData object with n_obs × n_vars = 4142 × 16106\n",
" obs: 'n_counts'\n",
" var: 'highly_variable', 'means', 'dispersions', 'dispersions_norm'\n",
" uns: 'hvg', 'pca'\n",
" obsm: 'X_pca'\n",
" varm: 'PCs'"
]
},
"execution_count": 7,
"metadata": {},
"output_type": "execute_result"
}
],
"source": [
"ad"
]
},
{
"cell_type": "markdown",
"metadata": {},
"source": [
"### Diffusion maps"
]
},
{
"cell_type": "markdown",
"metadata": {},
"source": [
"Palantir next determines the diffusion maps of the data as an estimate of the low dimensional phenotypic manifold of the data."
]
},
{
"cell_type": "code",
"execution_count": 8,
"metadata": {
"ExecuteTime": {
"end_time": "2022-06-15T17:45:43.083093Z",
"start_time": "2022-06-15T17:45:37.041390Z"
},
"execution": {
"iopub.execute_input": "2023-11-07T00:43:45.470863Z",
"iopub.status.busy": "2023-11-07T00:43:45.470133Z",
"iopub.status.idle": "2023-11-07T00:44:57.062600Z",
"shell.execute_reply": "2023-11-07T00:44:57.061672Z",
"shell.execute_reply.started": "2023-11-07T00:43:45.470823Z"
},
"tags": []
},
"outputs": [],
"source": [
"# Run diffusion maps\n",
"dm_res = palantir.utils.run_diffusion_maps(ad, n_components=5)"
]
},
{
"cell_type": "markdown",
"metadata": {},
"source": [
"The low dimensional embeddeing of the data is estimated based on the eigen gap using the following function"
]
},
{
"cell_type": "code",
"execution_count": 9,
"metadata": {
"ExecuteTime": {
"end_time": "2022-06-15T17:45:43.089195Z",
"start_time": "2022-06-15T17:45:43.085412Z"
},
"execution": {
"iopub.execute_input": "2023-11-07T00:44:57.063847Z",
"iopub.status.busy": "2023-11-07T00:44:57.063525Z",
"iopub.status.idle": "2023-11-07T00:44:57.067853Z",
"shell.execute_reply": "2023-11-07T00:44:57.067190Z",
"shell.execute_reply.started": "2023-11-07T00:44:57.063832Z"
},
"tags": []
},
"outputs": [],
"source": [
"ms_data = palantir.utils.determine_multiscale_space(ad)"
]
},
{
"cell_type": "markdown",
"metadata": {},
"source": [
"If you are specifying the number of eigen vectors manually in the above step, please ensure that the specified parameter is > 2"
]
},
{
"cell_type": "markdown",
"metadata": {},
"source": [
"### Visualization"
]
},
{
"cell_type": "markdown",
"metadata": {},
"source": [
"In the manuscript, we used tSNE projection using diffusion components to visualize the data. We now recommend the use of force-directed layouts for visualization of trajectories. Force-directed layouts can be computed by the same adaptive kernel used for determining diffusion maps.\n",
"\n",
"`scanpy` can be used to compute force directed layouts. We recommened the use of the diffusion kernel (see below) for computing force directed layouts\n",
"`Force Atlas` package should be installed for this analysis and can be installed using `conda install -c conda-forge fa2`. Note that fa2 is not supported by python3.9\n",
"\n",
"UMAPs are a good alternative to visualize trajectories in addition to force directed layouts."
]
},
{
"cell_type": "code",
"execution_count": 10,
"metadata": {
"ExecuteTime": {
"end_time": "2022-06-15T17:45:49.242486Z",
"start_time": "2022-06-15T17:45:43.091291Z"
},
"execution": {
"iopub.execute_input": "2023-11-07T00:44:57.068877Z",
"iopub.status.busy": "2023-11-07T00:44:57.068612Z",
"iopub.status.idle": "2023-11-07T00:46:13.903275Z",
"shell.execute_reply": "2023-11-07T00:46:13.902012Z",
"shell.execute_reply.started": "2023-11-07T00:44:57.068861Z"
},
"tags": []
},
"outputs": [],
"source": [
"sc.pp.neighbors(ad)\n",
"sc.tl.umap(ad)"
]
},
{
"cell_type": "code",
"execution_count": 11,
"metadata": {
"ExecuteTime": {
"end_time": "2022-06-15T17:45:49.341634Z",
"start_time": "2022-06-15T17:45:49.246116Z"
},
"execution": {
"iopub.execute_input": "2023-11-07T00:46:13.904867Z",
"iopub.status.busy": "2023-11-07T00:46:13.904495Z",
"iopub.status.idle": "2023-11-07T00:46:13.984173Z",
"shell.execute_reply": "2023-11-07T00:46:13.983323Z",
"shell.execute_reply.started": "2023-11-07T00:46:13.904850Z"
},
"tags": []
},
"outputs": [
{
"data": {
"image/png": "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",
"text/plain": [
"