Using multi-scale genomics to associate poorly annotated genes with rare diseases

Area of Science:

• Genomics
• Bioinformatics
• Computational Biology

Background:

• Next-generation sequencing (NGS) has advanced genetic disorder research but leaves many patients undiagnosed.
• Undiagnosed cases stem from hard-to-detect variants and mutations in novel disease genes.
• Identifying disease-causing genes requires sophisticated analytical tools.

Purpose of the Study:

• To introduce EvORanker, an algorithm for linking mutated genes to clinical phenotypes.
• To improve the diagnosis of genetic disorders by prioritizing candidate disease genes.
• To provide a user-friendly web tool for gene prioritization.

Main Methods:

• EvORanker integrates clinical data, multi-scale phylogenetic profiling, and omics data.
• The algorithm was validated using solved exomes and simulated genomes.
• Comparative analysis with existing methods was performed.

Main Results:

• EvORanker identified the correct disease gene in 69% of top candidates and 95% within the top 5.
• The algorithm demonstrated superior performance for poorly annotated genes compared to existing methods.
• EvORanker successfully identified candidate genes in previously unsolved genetic syndromes.

Conclusions:

• Clade-based phylogenetic profiling is an effective strategy for prioritizing disease genes.
• EvORanker demonstrates high efficacy in associating poorly annotated genes with patient phenotypes.
• The EvORanker web tool is publicly accessible for research use.