Evaluating phenotype-driven approaches for genetic diagnoses from exomes in a clinical setting

Area of Science:

• Genomics
• Clinical Medicine
• Bioinformatics

Background:

• Next-generation sequencing (NGS) advances molecular diagnostics for genetic conditions.
• Interpreting complex genetic variations remains a challenge.
• Variant prioritization methods integrate phenotype and genotype data.

Purpose of the Study:

• To compare the effectiveness of different variant prioritization methods for clinical exome analysis.
• To evaluate the performance of the PhenIX program against alternative tools.

Main Methods:

• A case series of 21 exomes with previously established causal variants was analyzed.
• Results from the PhenIX program were compared with five alternative methods.
• Variant prioritization performance was assessed by the rank of the true causal variant.

Main Results:

• PhenIX ranked the causal variant within the top 10 in 85% of cases.
• This performance was significantly better than the combined results of alternative methods (p=0.003).
• Exomiser (hiPHIVE) was the next best, ranking the causal variant in the top 10 in 25% of cases.

Conclusions:

• PhenIX demonstrates superior effectiveness for molecular diagnosis using clinical exome data.
• The distinct focuses of PhenIX (clinical, Mendelian genes) and Exomiser (gene discovery) impact their utility.
• Understanding these differences is crucial for selecting appropriate tools for molecular diagnosis.