The chief dawgs
We investigated whether a genome-scale CD4+ T-cell Perturb-seq screen contains reproducible regulatory architecture beyond ranked hits, statistical noise and connectivity hubs We built a quality-audited regulator × gene
Core trick
Spectral denoising of a CRISPRi Perturb-seq regulator-by-gene matrix, then consensus community detection checked against a matched null.
Tags
Description
We investigated whether a genome-scale CD4+ T-cell Perturb-seq screen contains reproducible regulatory architecture beyond ranked hits, statistical noise and connectivity hubs
We built a quality-audited regulator × gene × condition operator from a 1.8 TB public CRISPRi atlas using a laptop-scale pipeline. Spectral denoising reduced the operator to 92 empirically supported signal directions, and consensus community detection revealed structure 259 standard deviations above a matched null
The strongest result was annotation-blind recovery of mitochondrial Complex I, identified only after clustering, at BH-FDR 1.4 × 10⁻⁷. A second SAGA-centered module was supported by convergent evidence across CORUM, tensor factors and an independent K562 screen, while cross-cell-type analysis across CD4, K562 and RPE1 separated universal from T-cell-specific regulation
We also tested the predictive boundary of this structure. Real and shuffled regulator features performed identically for leave-regulator-out prediction, showing that recoverable structure does not imply inductive predictability
The result is an open, reproducible analysis and manuscript that others can build on
cuentadesanti/cd4-perturbseq-regulator-atlas★ 00 forksPythonMITCreated Jul 7, 2026 · last push Jul 11, 2026
README excerpt
# Robust regulators of CD4+ T cell programs — Perturb-seq > Hackathon submission · genome-scale CRISPRi Perturb-seq in primary human CD4+ T cells (Marson Lab, 2025) ## Hackathon submission — Built with Claude: Life Sciences - 🎬 **[Watch the 3-minute demo](https://youtu.be/3tu_3JmHh14)** - 📄 [Read the manuscript (PDF)](docs/CD4_operator_manuscript.pdf) - 🖼️ [View the flagship figure — Figure 3, operator modules](docs/figures/38_operator_insignia_3106.png) - 🔁 [Reproduce the analysis](#how-to-reproduce)  ## The question Which genes are **robust regulators** of CD4+ T cell programs? The challenge: the signal from a genome-scale screen is dominated by noise and by a handful of hubs. We want to **separate signal from noise with uncertainty** and prioritize regulators by a **large and reproducible** effect — not by raw counts or `adj_p_value < 0.1`. ## What this repo does This project turns a massive CD4+ T cell Perturb-seq screen into **three explorable objects: robust regulators, reproducibility audits, and transcriptional programs.** The CSV-only core identifies quality-aware regulators; optional h5ad slicing maps selected perturbations into transcriptomic fingerprint neighborhoods. It is **memory- and compute-aware** (the full dataset is 1.8 TB; the working disk has ~10 GB free) — the entire core runs **from the supplementary CSV tables alone (~15 MB)**: 1. **Selective download pipeline** from the public S3 bucket (`scr