crispyx

Motivation

Genome-wide CRISPR screens routinely produce datasets with hundreds of thousands of cells and tens of thousands of genes. Standard single-cell analysis toolkits (Scanpy, Pertpy) load the entire count matrix into memory, which can require 30–100+ GB of RAM and makes many screens impractical to analyse on commodity hardware or shared HPC nodes with per-job memory limits.

crispyx solves this by streaming data directly from on-disk AnnData (.h5ad) files. Quality control, normalisation, pseudo-bulk aggregation, and differential expression all operate without materialising the full matrix in memory, so even the largest screens can be processed with modest resources.

Features

• Streaming QC & preprocessing – Filter cells, perturbations, and genes; normalise and log-transform; all without loading the full matrix into memory
• Pseudo-bulk aggregation – Average log expression and pseudo-bulk count matrices for effect size estimation
• Differential expression – t-test, Wilcoxon rank-sum, and negative binomial GLM with apeGLM LFC shrinkage; multi-core support and adaptive memory management
• Dimension reduction – Memory-efficient PCA and KNN graph construction on backed data
• Scanpy-compatible API & plotting – Familiar cx.pp, cx.pb, cx.tl, and cx.pl namespaces; Scanpy-style rank genes plots, volcano, MA, PCA, UMAP, QC summaries, and overlap heatmaps
• Data preparation utilities – Edit backed metadata without loading X; standardise gene names; normalise perturbation labels; auto-detect metadata columns
• HPC-ready – Resume/checkpoint for long-running jobs; configurable memory_limit_gb; Docker and Singularity support

Quick Start

import crispyx as cx

# Open dataset without loading into memory
adata = cx.read_h5ad_ondisk("data/demo_benchmark.h5ad")

# Quality control with adaptive thresholds
adata = cx.pp.qc_summary(
    adata,
    perturbation_column="perturbation",
    min_genes=5,
    min_cells_per_perturbation=5,
)

# Differential expression
adata = cx.tl.rank_genes_groups(
    adata,
    perturbation_column="perturbation",
    method="wilcoxon",  # or "t-test", "nb_glm"
)

# Access results
print(adata.uns["rank_genes_groups"])
de_results = adata.uns["rank_genes_groups"].load()

For the full workflow (normalisation, PCA, pseudo-bulk, NB-GLM, LFC shrinkage, plotting, data preparation utilities), see the Usage Guide and the tutorial notebook.

Performance

Benchmarked across 12 CRISPR screen datasets (21k–1.97M cells), crispyx consistently outperforms Scanpy, Pertpy/PyDESeq2, and edgeR in both speed and memory:

Metric	crispyx vs Scanpy	crispyx vs Pertpy/PyDESeq2
t-test	2–11× faster	—
Wilcoxon	2–43× faster	—
NB-GLM	—	2× faster, completes where Pertpy OOMs
Peak memory	2–6× lower	Runs within 64 GB where Pertpy exceeds 120 GB
Accuracy	Pearson r > 0.999 vs Scanpy	Pearson r > 0.97 vs PyDESeq2

crispyx succeeds on all 12 datasets, while Scanpy times out or OOMs on the largest screens and Pertpy/edgeR fail on most genome-wide datasets.

See benchmarking/ for full results and reproduction scripts.

Installation

pip install -e .

Benchmarking

cd benchmarking
./run_benchmark.sh config/Adamson.yaml       # single dataset
./run_benchmark.sh config/*.yaml             # all datasets

See benchmarking/README.md for configuration options and output structure.

Testing

pytest

Documentation

sphinx-build docs docs/_build

Acknowledgements

crispyx builds on the foundational work of Scanpy (Wolf et al., 2018), Pertpy, PyDESeq2 (Muzellec et al., 2023), and AnnData (Virshup et al., 2024). We gratefully acknowledge these projects for establishing the single-cell analysis ecosystem in Python; crispyx extends their APIs and algorithmic designs to enable memory-efficient, streaming computation for large-scale CRISPR screen datasets.

Contributing

Suggestions, bug reports, and contributions are welcome! Please open an issue or submit a pull request.