Retrieval-augmented protein encoders that fuse homologous sequences into a single-pass transformer for variant effect and contact prediction.
Sparse mixture-of-experts autoregressive protein language model family pretrained on 1.5 trillion amino acid tokens with compute-optimal scaling.
Profluent / University of Washington / Massachusetts General Hospital / Harvard Medical School / Harvard University
Released January 6, 2025
CRISPR-Cas PAM specificity prediction directly from Cas protein sequence, plus computational evolution of Cas9 variants toward a chosen PAM.
Conditional adapters over a frozen ProGen2 generate enzymes steered by EC number, taxonomy, or free-text function, including rare and unseen classes.
AI-designed CRISPR-Cas9 gene editor generated by protein language models trained on 1.2 million CRISPR operons and shown to edit the human genome.