Bi-directional SIFT predicts a subset of activating mutations

Area of Science:

• Genomics
• Bioinformatics
• Computational Biology

Background:

• High-throughput sequencing generates vast amounts of genetic variation data.
• Identifying functional mutations, especially those causing hyperactivation or gain-of-function, is crucial but challenging.
• Existing computational tools lack specificity for gain-of-function mutation prediction.

Purpose of the Study:

• To develop a computational tool for identifying gain-of-function mutations.
• To adapt the Sorting Intolerant from Tolerant (SIFT) algorithm for enhanced functional prediction.
• To analyze large-scale cancer genotyping data for potential activating mutations.

Main Methods:

• Modified the SIFT algorithm to create bi-directional SIFT (B-SIFT).
• Utilized protein sequence alignments with homologous sequences to assess evolutionary fitness.
• Applied B-SIFT to experimental datasets and cancer genotyping data.

Main Results:

• B-SIFT successfully identified experimentally verified activating mutants.
• Analysis of cancer data revealed potential activating mutations.
• Structural evidence was provided for some identified mutations.

Conclusions:

• B-SIFT is effective in identifying gain-of-function mutations.
• The tool can aid in protein engineering and cancer mutation discovery.
• B-SIFT offers a valuable approach for functional mutation analysis.