Exponential Family Measurement Error Models for Single-Cell CRISPR Screens

• Activity Number: 151 - Novel Methods and Tools in the Era of Big Omics Data
• Type: Contributed
• Date/Time: Monday, August 8, 2022 : 10:30 AM to 12:20 PM
• Sponsor: Section on Statistics in Genomics and Genetics
• Author(s): Timothy Barry* and Eugene Katsevich and Kathryn Roeder
• Companies: Carnegie Mellon University and University of Pennsylvania and Carnegie Mellon University
• Keywords: CRISPR; single-cell; measurement error; exponential family; GLM; cloud

Abstract

CRISPR genome engineering and single-cell sequencing have transformed biological discovery. Single-cell CRISPR screens unite these two technologies, linking genetic perturbations in individual cells to changes in gene expression. 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 tuning parameter. To overcome these limitations, 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 response distributions and sources of measurement error 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 tens or hundreds of nodes on clouds (e.g., Microsoft Azure). Leveraging this infrastructure, we apply GLM-EIV to analyze two recent, single-cell CRISPR screen datasets.