A side-by-side comparison of variant function measurements using deep mutational scanning and base editing.

Area of Science:

• Mammalian functional genomics
• CRISPR base editing applications
• High-throughput screening methodologies

Background:

• Variant annotation is critical for understanding gene function in mammals.
• Deep mutational scanning (DMS) using cDNA libraries is a standard method.
• CRISPR base editing (BE) offers a promising alternative for variant annotation.

Purpose of the Study:

• To directly compare CRISPR base editing (BE) with cDNA deep mutational scanning (DMS) for variant annotation.
• To assess the reliability of high-throughput BE for annotating variant function.
• To determine the necessary extent of downstream experimental validation for BE screens.

Main Methods:

• Direct comparison of cDNA DMS and BE in the same laboratory and cell line.
• Analysis of short guide RNA (sgRNA) depletion/enrichment based on predicted edits within the sgRNA's 'editing window'.
• Application of filters for single and multiple edits to assess variant annotation accuracy.

Main Results:

• High correlation observed between base editor data and gold-standard DMS data.
• Predicted edits within the sgRNA window significantly improve agreement between DMS and BE.
• A simple filter for sgRNAs with single edits in their window can annotate a large proportion of variants.
• Measuring edits in medium-sized validation pools recovers high-quality data for multi-edit guides.

Conclusions:

• CRISPR base editing screens show a high degree of correlation with DMS, validating its utility for variant annotation.
• Simple filtering strategies can enable direct variant annotation from BE screens, potentially reducing validation needs.
• The primary measurable variable in base editor screens is the intended base edits, supporting their functional relevance.