Protein sequence‐to‐structure learning: Is this the end(‐to‐end revolution)?. Issue 12 (22nd September 2021)

• Record Type: Journal Article
• Authors: Laine, Elodie; Eismann, Stephan; Elofsson, Arne; Grudinin, Sergei
• Other Names: Moult John guestEditor.; Kryshtafovych Andriy guestEditor.
• Is Part Of: Proteins. Volume 89:Issue 12(2021)
• Page Start: 1770
• Page End: 1786
• Publication Date: 2021-09-22
• Subjects: CASP14 -- deep learning -- end‐to‐end architectures -- equivariance -- geometric learning -- protein language models -- protein structure prediction; Proteins -- Periodicals; Proteins -- Periodicals; 572.6
• DOI: 10.1002/prot.26235
• Languages: English
• ISSNs: 0887-3585

Abstract

Abstract: The potential of deep learning has been recognized in the protein structure prediction community for some time, and became indisputable after CASP13. In CASP14, deep learning has boosted the field to unanticipated levels reaching near‐experimental accuracy. This success comes from advances transferred from other machine learning areas, as well as methods specifically designed to deal with protein sequences and structures, and their abstractions. Novel emerging approaches include (i) geometric learning, that is, learning on representations such as graphs, three‐dimensional (3D) Voronoi tessellations, and point clouds; (ii) pretrained protein language models leveraging attention; (iii) equivariant architectures preserving the symmetry of 3D space; (iv) use of large meta‐genome databases; (v) combinations of protein representations; and (vi) finally truly end‐to‐end architectures, that is, differentiable models starting from a sequence and returning a 3D structure. Here, we provide an overview and our opinion of the novel deep learning approaches developed in the last 2 years and widely used in CASP14.