Genome-wide prediction of disease variant effects with a deep protein language model

By Nadav Brandes, Grant Goldman, Charlotte H. Wang, Chun Jimmie Ye, Vasilis Ntranos. Nature Genetics, 2023

348 citations recorded by the source corpus.

Predicting the effects of coding variants is a major challenge. While recent deep-learning models have improved variant effect prediction accuracy, they cannot analyze all coding variants due to dependency on close homologs or software limitations. Here we developed a workflow using ESM1b, a 650-million-parameter protein language model, to predict all ~450 million possible missense variant effects in the human genome, and made all predictions available on a web portal. ESM1b outperformed existing methods in classifying ~150,000 ClinVar/HGMD missense variants as pathogenic or benign and predicting measurements across 28 deep mutational scan datasets. We further annotated ~2 million variants as damaging only in specific protein isoforms, demonstrating the importance of considering all isoforms when predicting variant effects. Our approach also generalizes to more complex coding variants such as in-frame indels and stop-gains. Together, these results establish protein language models as an effective, accurate and general approach to predicting variant effects. A modified framework leveraging a protein language model (ESM1b) is used to predict all possible 450 million missense variant effects in the human genome and shows potential for generalizing to more complex genetic variations such as indels and stop-gains.

Subfields: Genetics, Molecular Biology

View Paper: https://www.semanticscholar.org/paper/50d55d0ccdb32e43686727adac24c84f7ecf40db

How researchers see this paper

The researchers whose trusted networks rank it highest, strongest advocate first. Every researcher gets a different order.

In Genetics

1. Max Schubach
2. Lusiné Nazaretyan
3. Thorben Maass
4. Sebastian Röner
5. Martin Kircher

In Molecular Biology

1. Max Schubach
2. Lusiné Nazaretyan
3. Thorben Maass
4. Sebastian Röner
5. Martin Kircher

Citations in the corpus

• CADD v1.7: using protein language models, regulatory CNNs and other nucleotide-level scores to improve genome-wide variant predictions (2024, methodology): “It should also be noted that distinguishing methods by their means of training only (e.g. supervised vs. unsupervised learning approaches) is insufficient and that the training objective needs to be considered ( 46 ,47 ).”