Pangu-Weather

AntiFold — antibody-specific inverse folding model built on ESM-IF1, fine-tuned on antibody structures from SAbDab and OAS databases. Predicts residue log-likelihoods for antibody variable domains (IM

1 shared topic • 3 shared operations

GNNExplainer explains predictions made by graph neural networks by learning sparse masks over subgraph structure and node features. Use for post-hoc GNN interpretability, identifying important edges a

1 shared topic • 3 shared operations

Use when working with NetMHCpan 4.1 for MHC class I peptide binding and ligand presentation prediction. NetMHCpan accepts peptide lists or FASTA proteins, scores candidate ligands against up to 20 all

1 shared topic • 3 shared operations

Use when working with Sapiens — a BERT-based human antibody language model for antibody humanization, humanness scoring, and sequence design. Sapiens predicts humanizing mutations, scores residue-leve

1 shared topic • 3 shared operations

Boltz-1 — open-source deep learning model for predicting biomolecular 3D structures and interactions, approaching AlphaFold3-level accuracy. Supports protein, DNA, RNA, and small-molecule ligand struc

1 shared topic • 2 shared operations