D. Akpinaroglu

Deep learning guided design of dynamic proteins

This paper designs dynamic proteins with switch-like behavior using deep learning, validated by structure solving and simulations showing controllable conformational changes.

Structure-conditioned masked language models for protein sequence design generalize beyond the native sequence space

This paper introduces Frame2seq, a structure-conditioned masked language model for protein sequence design that generates novel, stable proteins with low sequence identity to native sequences.

Separable mechanisms drive local and global polarity establishment in the C. elegans intestinal epithelium

The paper shows that local and global polarity in C. elegans intestinal epithelium are temporally and genetically separable, with distinct roles for PAR-3 and PKC-3/aPKC.

Improved antibody structure prediction by deep learning of side chain conformations

This paper introduces DeepSCAb, a deep learning method that improves antibody structure prediction by modeling side chain conformations, outperforming existing methods that lack side chain context.