PRESCOTT: a population aware, epistatic, and structural model accurately predicts missense effects

Area of Science:

• Genomics
• Bioinformatics
• Computational Biology

Background:

• Predicting the functional consequences of genetic point mutations is a significant challenge in genomics.
• Accurate variant classification is crucial for understanding genetic diseases and advancing genomic medicine.

Purpose of the Study:

• To develop and validate PRESCOTT, a novel computational tool for predicting the functional impact of missense variants.
• To reconstruct comprehensive mutational landscapes and identify mutation-sensitive protein regions.

Main Methods:

• PRESCOTT integrates protein sequence data, structural models, and population-specific allele frequencies.
• The method reconstructs mutational landscapes and categorizes variants as benign, pathogenic, or of uncertain significance.

Main Results:

• PRESCOTT demonstrates superior performance in classifying variants from ClinVar and the ACMG dataset compared to existing methods.
• The tool successfully analyzed over 1800 proteins from the Human Protein Dataset, identifying mutation-sensitive regions.

Conclusions:

• PRESCOTT offers a robust platform for predicting mutation effects across diverse proteins and variants.
• The tool's online server and open-access data provide valuable resources for researchers in genomic medicine and variant interpretation.