Increasing the fraction of correct solutions in ensembles of protein-protein docking models by an iterative consensus algorithm.
Increasing the fraction of correct solutions in ensembles of protein-protein docking models by an iterative consensus algorithm.
- June 14, 2026
- Posted by: Stem Skills Lab
Protein Science · 2025
Tiziana Ricciardelli, Didier Barradas‐Bautista, Zhen Cao, Mohit Chawla, Luigi Cavallo, Romina Oliva
Protein–protein docking programs produce huge ensembles of candidate structures, and picking the correct one is hard. This study introduces Iter-CONSRANK, an iterative consensus-scoring method that repeatedly filters out lower-ranked models based on conserved residue contacts. Tested on datasets of up to millions of models, it pushed correct solutions into the top ranks, outperforming over 150 other scoring functions and enriching correct models up to eightfold for difficult targets.
Learn these methods hands-on
This is the kind of computational work the StemSkills team publishes. Our live Molecular Modeling & MD Simulations cohort teaches you to do it yourself, mentored by published researchers.
See the bootcamp →