Gustav Oberdorfer

This study modifies de novo α-helical barrel biocatalysts using deep learning and computational methods to improve solubility and catalytic performance while maintaining structural stability.

This paper introduces Riff-Diff, a machine learning and atomistic modeling method for designing enzymes with high catalytic activity and stereoselectivity, demonstrating its effectiveness for two chemical reactions.

The paper introduces ELEN, a model for assessing protein loop quality in structure predictions, showing it can identify poor design positions in redesigned enzymes.

The paper engineers transmembrane alkane monooxygenases to improve terminal hydroxylation of isoprenyl acetate, enabling fermentative production of tulipalin A from renewable resources.

The paper designs and characterizes a thermostable α-helical barrel protein with tunable active-site positioning, showing improved catalytic efficiency.

This paper improves prediction performance on small DMS datasets by using structural representations and data augmentation, achieving results comparable to sequence-based methods on large datasets.