Huazhong Agricultural University

Plant genome foundation model for ab initio gene structure annotation, predicting genes, coding sequences, and exons at single-nucleotide resolution.

Multimodal pre-training distils 3D, text and biochemical knowledge into a 2D molecular graph encoder, so downstream prediction needs only SMILES.

Drug-drug interaction event text generation from two molecular structures, conditioned on biological functions selected for each drug in the pair.

Plant genomic foundation model with a 64 kb single-nucleotide context that predicts gene structures and generates de novo plant gene sequences.