Exponential family measurement error models for single-cell CRISPR screens

TL;DR

This paper introduces GLM-EIV, a novel statistical method for analyzing single-cell CRISPR screens that addresses biases in standard approaches, enabling more accurate inference of gene regulation from noisy, high-dimensional data.

Contribution

The paper develops GLM-EIV, extending errors-in-variables models to exponential family responses, and provides scalable computational tools for large-scale single-cell CRISPR data analysis.

Findings

GLM-EIV reduces bias compared to thresholded regression.

Application to real datasets reveals new biological insights.

Scalable infrastructure enables analysis of large-scale screens.

Abstract

CRISPR genome engineering and single-cell RNA sequencing have accelerated biological discovery. Single-cell CRISPR screens unite these two technologies, linking genetic perturbations in individual cells to changes in gene expression and illuminating regulatory networks underlying diseases. Despite their promise, single-cell CRISPR screens present substantial statistical challenges. We demonstrate through theoretical and real data analyses that a standard method for estimation and inference in single-cell CRISPR screens -- "thresholded regression" -- exhibits attenuation bias and a bias-variance tradeoff as a function of an intrinsic, challenging-to-select tuning parameter. To overcome these difficulties, we introduce GLM-EIV ("GLM-based errors-in-variables"), a new method for single-cell CRISPR screen analysis. GLM-EIV extends the classical errors-in-variables model to responses and noisy predictors that are exponential family-distributed and potentially impacted by the same set of confounding variables. We develop a computational infrastructure to deploy GLM-EIV across hundreds of processors on clouds (e.g., Microsoft Azure) and high-performance clusters. Leveraging this infrastructure, we apply GLM-EIV to analyze two recent, large-scale, single-cell CRISPR screen datasets, yielding several novel insights.