GDIv2: improving variant selection from human exomes.

Area of Science:

• Genetics
• Bioinformatics
• Genomic Medicine

Background:

• The Gene Damage Index (GDI) is used to assess cumulative mutational damage in genes.
• The original GDI has limitations, including bias from coding sequence length and lack of gene-specific variant deleteriousness.
• Accurate prioritization of candidate disease genes is crucial for genetic studies.

Purpose of the Study:

• To develop an improved framework, GDIv2, for quantifying gene damage.
• To correct for coding sequence length and incorporate gene-specific normalization of variant deleteriousness scores (CADD).
• To enhance the discrimination between disease-relevant and non-relevant genes.

Main Methods:

• Developed GDIv2 by normalizing CADD scores and correcting for coding sequence length.
• Generated four GDIv2 implementations using 1000 Genomes Project and gnomAD datasets on GRCh37 and GRCh38 genome builds.
• Benchmarked GDIv2 against the original GDI and other methods like RVIS, LOEUF, s_het, and CoNeS.

Main Results:

• All GDIv2 versions significantly improved discrimination between relevant and accessory genes compared to the original GDI.
• GDIv2 reduced the erroneous exclusion of relevant genes and increased the exclusion of accessory genes.
• GDIv2_1kGP_37 achieved high performance, excluding 24.6% of accessory genes while retaining 96.7% of relevant genes; combining it with other tools further improved filtering.

Conclusions:

• GDIv2 offers a more accurate and robust method for assessing gene damage and prioritizing candidate disease genes.
• The updated framework enhances the reliability of genetic studies by improving gene filtering.
• GDIv2 resources are publicly available for researchers.