What question did this study set out to answer?

To develop a machine learning framework that predicts Mendelian genes and uncover what distinguishes them from non-Mendelian genes.

February 19, 2026PLoS Computational BiologyOpen Access

MENDELSEEK: An algorithm that predicts mendelian genes and elucidates what makes them special

View Full Paper

Ask AI

Bookmark

Authors

MENDELSEEK reveals 1,277 novel Mendelian gene candidates in humans, highlighting their unique biochemical characteristics.

To develop a machine learning framework that predicts Mendelian genes and uncover what distinguishes them from non-Mendelian genes.
Introduced MENDELSEEK machine learning algorithm
Integrated residue variation scores with pathway participation
Utilized Gene Ontology processes and protein language model features
Performed benchmarking with 10-fold cross-validation on 16,946 human genes
Achieved AUC of 0.869 and AUPR of 0.737 in performance metrics
Outperformed methods ENTPRISE+ENTPRISE-X and REVEL
Predicted 1,277 novel Mendelian gene candidates with precision greater than 0.7
Mendelian genes showed significantly more protein-protein interactions than non-Mendelian genes

Zhou et al. (2026) studied this question.

View Full Paper

Ask AI

Bookmark