University of Adelaide

Bacteriophage gene function prediction from genomic synteny, pairing protein language model embeddings with circular attention. AUC above 0.84.

RNA inverse folding model that designs nucleotide sequences for a target 3D backbone by running discrete diffusion in hyperbolic space.

Kinase-inhibitor binding affinity prediction fusing a contrastively pretrained molecular graph encoder with structure-informed kinase embeddings.

Predicts binding free energy change (ΔΔG) at protein-protein interfaces by scoring bound and unbound states with an inverse folding model.

3D CT segmentation of abdominal organs and tumors, where a deformable-attention Transformer decodes organ embeddings into the segmentation kernels.

Multi-organ and tumor segmentation in 3D abdominal CT from a single network whose segmentation kernels are generated per task by a controller.