Julian M. Hahnfeld

This study compares deep learning models for generating antimicrobial peptides, finding that auto-encoders, especially Wasserstein auto-encoders, produce more diverse sequences than other models.

The paper identifies uORFs in bacterial mRNAs using bioinformatics and experimental validation, revealing translational coupling and ribosome-mediated RBS occlusion as regulatory mechanisms.

sORFdb is a database for small proteins and sORFs in bacteria, addressing gaps in detection and functional annotation through curated data and family classification.