TEV Protease - Homolog and SSVL Libraries

The project aims to collect data on thousands of protease sequence variants, with applications in understanding protease specificity and developing therapeutic interventions.

This dataset combines a sequence-diverse TEV protease homolog library (4,022 natural homologs plus 422 AI-generated variants from the SCISOR discrete-diffusion model) with a site-saturation variant library (SSVL) of TEV. Assayed via split-DHFR at the Living Measurement Systems Foundry (LMSF) at NIST.