Massively parallel quantification and prediction of monomeric protein yield in Escherichia coli

A pooled assay in Escherichia coli quantifies monomeric yield for over 39,000 protein domain variants, enabling sequence-based prediction and candidate enrichment.

• Why it matters: Reliable, standardized monomeric-yield measurements are scarce, limiting identification of determinants and the development of accurate ranking predictors for de novo designs and binders.
• What they did: The team combined peptide barcodes, size-exclusion chromatography, and mass spectrometry under uniform Escherichia coli production, then trained ESMC-MonoYield to predict monomeric yield from amino-acid sequence.
• The result: Folding stability, inverse-folding benchmarks, structure-prediction confidence metrics, and nanobody sequence/structure patterns were linked to monomeric yield, and ESMC-MonoYield prospectively enriched high-yield BindCraft-designed binders.

AI summary of the abstract · may contain errors · not medical advice · read the paper.