Dmitrij Frishman

BLMPred: predicting linear B-cell epitopes using pre-trained protein language models and machine learning

BLMPred predicts linear B-cell epitopes using pre-trained protein language models and machine learning, outperforming existing tools on multiple datasets.

EF-P and its paralog EfpL (YeiP) differentially control translation of proline containing sequences

This study identifies EfpL as a ribosome rescue factor that senses cellular metabolic state, complementing EF-P in proline-containing sequence translation.

Prognostic importance of splicing-triggered aberrations of protein complex interfaces in cancer

The paper identifies prognostic exon-exon interactions in cancer by analyzing RNA-seq and 3D protein structures, showing their association with patient survival.

Deep learning model can predict water binding sites on the surface of proteins using limited-resolution data

The paper introduces a deep learning model called hotWater to predict water binding sites on protein surfaces using limited-resolution data, showing improved performance over existing methods.