Model-based understanding of single-cell CRISPR screening.

Area of Science:

• Genomics
• Molecular Biology
• Bioinformatics

Background:

• Single-cell CRISPR screening techniques like Perturb-Seq, CRISP-seq, and CROP-seq integrate pooled CRISPR screening with single-cell RNA sequencing.
• These methods enable functional genetic screens at single-cell resolution, offering unprecedented insights into gene function.

Purpose of the Study:

• To introduce MUSIC, an integrated computational pipeline designed for model-based analysis of single-cell CRISPR screening data.
• To provide a robust method for understanding gene perturbation effects and biological circuits from complex single-cell screening datasets.

Main Methods:

• Developed MUSIC, an integrated pipeline for analyzing single-cell CRISPR screening data.
• Validated MUSIC using publicly available datasets to assess its performance in quantifying and prioritizing gene perturbation effects.
• Tested MUSIC's ability to handle noise inherent in single-cell RNA sequencing data.

Main Results:

• MUSIC accurately quantifies and prioritizes individual gene perturbation effects on cell phenotypes.
• The pipeline demonstrates tolerance for substantial noise commonly found in single-cell CRISPR screening data.
• MUSIC effectively analyzes perturbation effects from overall impact to functional topic-specific and inter-perturbation relationships.

Conclusions:

• MUSIC offers an effective and applicable solution for elucidating perturbation function and biological circuits.
• The model-based quantitative analysis provided by MUSIC enhances the interpretation of single-cell CRISPR screening data.
• This pipeline facilitates a deeper understanding of gene function and regulatory networks at the single-cell level.