Jan Maciej Zaucha

The paper identifies bimodal protein expression patterns in cancer proteomics, revealing 2401 tumor-associated bimodal proteins and their associations with key cancer pathways, with implications for biomarker discovery and precision oncology.

The paper introduces a machine learning model that learns spatial preference maps for drug-receptor interactions to guide drug discovery.

The paper presents a genetics-first framework using molecular knowledge to predict phenotypes from rare coding variants in a direct-to-consumer genotype cohort, identifying novel genetic causes for developmental disorders.

The paper develops a method to analyze the secretome of mouse brain cells and identifies cell type-specific secreted proteins, revealing proteolytic cleavage of membrane proteins.

The paper identifies an epistatic interaction between ACTN3 and SNAP-25 polymorphisms that improves identification of gymnastic aptitude compared to additive models.

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.