Unfolded and nonfunctional isoforms of folded proteins are incorrectly predicted as folded by protein folding neural networks

Summary

Unfolded and nonfunctional isoforms of folded proteins are predicted as folded by PLM-based protein folding neural networks (1). (2) first demonstrated that isoforms of well-folded proteins can be unstructured.

1.

Zhang Z, Wayment-Steele HK, Brixi G, Wang H, Kern D, Ovchinnikov S. Protein language models learn evolutionary statistics of interacting sequence motifs. Proceedings of the National Academy of Sciences. 2024;121(45). Available from: https://doi.org/10.1073/pnas.2406285121

2.

Pozo F, Martinez-Gomez L, Walsh TA, Rodriguez JM, Di Domenico T, Abascal F, et al. Assessing the functional relevance of splice isoforms. NAR Genomics and Bioinformatics. 2021;3(2). Available from: https://doi.org/10.1093/nargab/lqab044