Figure - PMC

Skip to main content

An official website of the United States government

Here's how you know

Here's how you know

Official websites use .gov
A .gov website belongs to an official government organization in the United States.

Secure .gov websites use HTTPS
A lock ( ) or https:// means you've safely connected to the .gov website. Share sensitive information only on official, secure websites.

View full-text article in PMC

. 2026 Mar 30;23(4):772–784. doi: 10.1038/s41592-026-03050-9

Search in PMC
Search in PubMed
View in NLM Catalog
Add to search

© The Author(s) 2026

Open Access This article is licensed under a Creative Commons Attribution 4.0 International License, which permits use, sharing, adaptation, distribution and reproduction in any medium or format, as long as you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons licence, and indicate if changes were made. The images or other third party material in this article are included in the article’s Creative Commons licence, unless indicated otherwise in a credit line to the material. If material is not included in the article’s Creative Commons licence and your intended use is not permitted by statutory regulation or exceeds the permitted use, you will need to obtain permission directly from the copyright holder. To view a copy of this licence, visit http://creativecommons.org/licenses/by/4.0/.

PMC Copyright notice

Extended Data Fig. 2 — (a) Performance of ESMIN on ClinVar (2,400 genes with a total of ~27k pathogenic and benign variants, class balanced per gene) compared to other ESM family ensembles and individual models. ESM_Family is composed of the 11 ESM models contributing to ESMIN. ESM1b + 2 is the equal-weight average of the ESM1b and ESM2-650M models and ESM1v is the ensemble of the 5 individual ESM1v models. (b) Performance evaluation on ProteinGym (120 organismal fitness and activity DMS assays, across 101 proteins). The top right panel shows the percentage of DMS assays for which an ESM ensemble outperforms all individual ESM family models. (c) Heatmap of Spearman correlations between ESM model predictions that contribute to ESMIN. (d) The fraction of model predictions included in ESMIN and the Spearman correlation with ESMIN are shown for each model. (e) Effect of adding ESM2-15B to the family; ESMIN and average ensemble performance on ClinVar (top) and ProteinGym DMS (bottom), showing negligible additional gains from including ESM2-15B.