About ESMFold
Evolutionary-Scale Protein Structure Prediction
Cite This Tool
If you use ESMFold in your research, please cite the original paper:
DOI: 10.1126/science.ade2574
BibTeX
@article{lin2023esmfold,
title={Evolutionary-scale prediction of atomic-level protein structure with a language model},
author={Lin, Zeming and Akin, Halil and Rao, Roshan and Hie, Brian and Zhu, Zhongkai and Lu, Wenting and Smetanin, Nikita and Verkuil, Robert and Kabeli, Ori and Shmueli, Yaniv and dos Santos Costa, Allan and Fazel-Zarandi, Maryam and Sercu, Tom and Candido, Salvatore and Rives, Alexander},
journal={Science},
volume={379},
number={6637},
pages={1123--1130},
year={2023},
publisher={American Association for the Advancement of Science},
doi={10.1126/science.ade2574}
}
License: MIT License — Open source, freely available for academic and commercial use. Developed by Meta AI (FAIR).
Overview
ESMFold predicts full atomic-level protein structure directly from a single amino acid sequence using the ESM-2 protein language model, without requiring multiple sequence alignments (MSAs).
It achieves accuracy competitive with AlphaFold2 and RoseTTAFold while being significantly faster for single-sequence predictions. The model was trained on evolutionary data at an unprecedented scale, enabling it to internalize structural patterns directly from protein sequences. This makes it ideal for rapid structure prediction, especially for orphan proteins with limited evolutionary information.
Key Capabilities
-
Single-Sequence Prediction
No MSA or template search required, enabling rapid inference
-
Atomic-Level Accuracy
Predicts full backbone and side-chain atom coordinates
-
Per-Residue Confidence (pLDDT)
Provides reliability scores for each residue position
-
Global Confidence (pTM)
Overall predicted template modeling score for the structure
Applications
-
Rapid Structure Screening
Quickly predict structures for large protein sets
-
Metagenomic Protein Characterization
Predict structures for novel proteins with no known homologs
-
Drug Target Analysis
Identify binding sites and structural features for drug design
-
Protein Engineering
Evaluate structural impact of mutations and sequence modifications