DiMSum

DiMSum estimates variant fitness and quantifies error directly from raw deep mutational scanning (DMS) sequencing data to analyze the effects of thousands of variants on proteins, RNAs, and regulatory elements.

Key Features:

• Variant fitness and error estimation: Computes variant fitness values and associated error estimates directly from raw sequencing data.
• Interpretable error model: Implements an interpretable error model that captures the primary sources of variability inherent in DMS experiments.
• Customizable end-to-end pipeline: Provides a customizable end-to-end pipeline for processing and analyzing DMS datasets.
• Summary reports: Generates summary reports that identify common DMS experimental pathologies and support diagnosis of data issues.
• Support for diverse target types: Applies to variant effect analysis for proteins, RNAs, and regulatory elements.

Scientific Applications:

• Variant effect mapping: Mapping functional effects of thousands of sequence variants on proteins, RNAs, and regulatory elements.
• Fitness landscape quantification: Quantifying variant fitness landscapes to assess functional impacts across large variant sets.
• DMS quality control: Diagnosing and quantifying experimental variability and sources of error in DMS datasets.

Methodology:

Processes raw sequencing data to produce variant fitness and error estimates using an interpretable error model within a customizable end-to-end pipeline and outputs summary reports.