Tel Aviv University

Generative transformer for phylogenetic inference that transduces sets of unaligned molecular sequences directly into Newick-format trees.

Generative transformer for ancestral protein sequence reconstruction that needs no multiple sequence alignment or phylogenetic tree as input.

Protein sequence-structure co-embedding model placing domains, full sequences, and short segments in one 32-dimensional contrastive space.

Generates free-text descriptions of protein function, catalytic activity, subcellular localization, and domains from sequence alone.

Spatial transcriptomics prediction from H&E slides, inferring spot-level expression and tumor microenvironment composition in breast cancer.

Word2vec-based language model trained on 360 million microbial genes that predicts gene function from genomic context without sequence homology.