Exécution de notre outil de synthèse sur un nouvel article
Vérifiez l’article ci-dessous, puis lancez sa synthèse.
Protein structure prediction using multiple deep neural networks in the 13th Critical Assessment of Protein Structure Prediction (CASP13)
Andrew Senior (Google DeepMind (United Kingdom)), Richard Evans (Google DeepMind (United Kingdom)), John Jumper (Google DeepMind (United Kingdom)), James Kirkpatrick (Google DeepMind (United Kingdom)), Laurent Sifre (Google DeepMind (United Kingdom)), Tim Green (Google DeepMind (United Kingdom)), Chongli Qin (Google DeepMind (United Kingdom)), Augustin Žídek (Google DeepMind (United Kingdom)), Alexander W. R. Nelson (Google DeepMind (United Kingdom)), Alex Bridgland (Google DeepMind (United Kingdom)), Hugo Penedones (Google DeepMind (United Kingdom)), Stig Petersen (Google DeepMind (United Kingdom)), Karen Simonyan (Google DeepMind (United Kingdom)), Steve Crossan (Google DeepMind (United Kingdom)), Pushmeet Kohli (Google DeepMind (United Kingdom)), David T. Jones (The Francis Crick Institute, University College London), David Silver (Google DeepMind (United Kingdom)), Koray Kavukcuoglu (Google DeepMind (United Kingdom)), Demis Hassabis (Google DeepMind (United Kingdom))
Résumé : We describe AlphaFold, the protein structure prediction system that was entered by the group A7D in CASP13. Submissions were made by three free-modeling (FM) methods which combine the predictions of three neural networks. All three systems were guided by predictions of distances between pairs of residues produced by a neural network. Two systems assembled fragments produced by a generative neural network, one using scores from a network trained to regress GDT_TS. The third system shows that simple gradient descent on a properly constructed potential is able to perform on par with more expensive traditional search techniques and without requiring domain segmentation. In the CASP13 FM assessors' ranking by summed z-scores, this system scored highest with 68.3 vs 48.2 for the next closest group (an average GDT_TS of 61.4). The system produced high-accuracy structures (with GDT_TS scores of 70 or higher) for 11 out of 43 FM domains. Despite not explicitly using template information, the results in the template category were comparable to the best performing template-based methods.