A framework for automated gene selection in genomic applications.

Area of Science:

• Genomics
• Bioinformatics
• Medical Genetics

Background:

• Genomic data evaluation requires efficient identification of disease-associated genes.
• Individuals with unknown disease etiology and those undergoing genomic screening benefit from such tools.

Purpose of the Study:

• To propose a framework for gene selection in genomic analyses.
• To support applications using genes with established or emerging disease evidence.

Main Methods:

• Compiled a comprehensive gene list (6,145 genes) from Human Gene Mutation Database, OMIM, and ClinVar.
• Applied stringent filters and computationally curated evidence (DisGeNET) to create a refined list (3,929 genes).

Main Results:

• The generated gene lists show high inclusion of genes with strong disease associations compared to manual curation.
• Limited-evidence genes were largely excluded, enhancing specificity.
• The framework successfully identified pathogenic variants in 45 genomes.

Conclusions:

• The developed approach efficiently generates highly sensitive gene lists for genomic applications.
• The framework is dynamic and updatable, offering significant time savings.
• This method aids in evaluating genomic data for disease association.