Back

Genomic heterogeneity inflates the performance of variant pathogenicity predictions

Lu, B.; Liu, X.; Lin, P.-Y.; Brandes, N.

2025-09-08 bioinformatics
10.1101/2025.09.05.674459 bioRxiv
Show abstract

Recent studies have reported unprecedented accuracy predicting pathogenic variants across the genome, including in noncoding regions, using large AI models trained on vast genomic data. We present a comprehensive evaluation of these frontier models, showing that performance is inflated by differences in the prevalence of pathogenic variants across genomic contexts. We identify the best-performing models for each variant type and establish a benchmark to guide future progress.

Matching journals

The top 6 journals account for 50% of the predicted probability mass.

50% of probability mass above

"Similar papers" are the closest papers from that journal in the model's embedding space. They show what the match is built on, but the ranking comes mostly from a classifier over the whole training set, not from these examples alone.