K. Anton Feenstra

PIPENN-EMB improves protein interface prediction using protein embeddings, achieving better performance and generalization to remote homologs.

The paper introduces ProteinGLUE, a benchmark suite for evaluating self-supervised protein models, showing pre-training improves downstream tasks but smaller models perform better than larger ones.

The paper introduces an ensemble of deep learning models (PIPENNs) for predicting protein interface residues, showing improved performance over existing methods across different interaction types.

The paper introduces SeRenDIP-CE, a sequence-based method for predicting conformational epitopes, achieving better performance than existing predictors on an antibody-antigen structure.

The paper compares genetic interaction patterns between gene-specific transcription factors and signaling genes in yeast, finding inversion is more common in transcription factors due to differential regulation.

This study constructs and validates a Petri net model of Wnt/β-catenin signaling to investigate its behavior under physiological and pathophysiological conditions, revealing dose- and time-dependent responses and higher pathway activity in hyperactive signaling.