University of Michigan

Multi-task cellular foundation model predicting drug sensitivity, perturbation expression and drug-protein binding from one pretrained checkpoint.

Drug- and dose-conditioned latent transition predictor pretrained on the Tahoe-100M perturbation atlas and transferred frozen to tumor RNA-seq.

Generative chemistry ensemble fine-tuned on 2.4M RNA-small molecule binding measurements, designing analogues around a seed compound.

Chromatin-informed foundation model predicting regulatory activity and chromatin state directly from plant genomic sequence in Arabidopsis and rice.

Foundation model for 3D genome architecture, using masked locus modeling over genome-wide contact profiles to capture chromosome-scale organization.

Generalist neuroimaging vision foundation model pretrained on 5.24M clinical MRI and CT volumes for radiologic diagnosis and report generation.

Enzyme Commission number prediction that pools ESM Cambrian embeddings across unlabeled sequence homologs, scoring 0.788 F1 on full 4-digit EC.

Viral protein language model that predicts a virus's animal host from one protein sequence, generalizing to rare and unseen hosts at 18M parameters.

Vaccine literature-mining model that classifies Brucella vaccine abstracts and extracts antigen, formulation, platform, and animal-model fields.

Cell Painting image generation conditioned on a control well image and a compound's structure, covering cell lines and chemicals never trained on.

Gut microbiome language model that reads a 16S sample as a sentence of taxa, producing context-sensitive embeddings that transfer across cohorts.