{"about":{"site":"https://codewithpapers.app","non_affiliation":"Code with Papers and Syntology are not affiliated with, endorsed by, or sponsored by Papers with Code, Meta, or the pwc-archive mirror.","licence":"CC BY-SA 4.0","licence_url":"https://creativecommons.org/licenses/by-sa/4.0/legalcode","attribution":"https://codewithpapers.app/attribution","modified":"archive material modified by Syntology; see the attribution page"},"url":"/paper/bayesian-active-learning-for-optimization-and","title":"Bayesian active learning for optimization and uncertainty quantification in protein docking","arxiv_id":"1902.00067","date":"2019-01-31","proceeding":null,"authors":["Yue Cao","Yang shen"],"abstract":"Motivation: Ab initio protein docking represents a major challenge for\noptimizing a noisy and costly \"black box\"-like function in a high-dimensional\nspace. Despite progress in this field, there is no docking method available for\nrigorous uncertainty quantification (UQ) of its solution quality (e.g.\ninterface RMSD or iRMSD).\n  Results: We introduce a novel algorithm, Bayesian Active Learning (BAL), for\noptimization and UQ of such black-box functions and flexible protein docking.\nBAL directly models the posterior distribution of the global optimum (or native\nstructures for protein docking) with active sampling and posterior estimation\niteratively feeding each other. Furthermore, we use complex normal modes to\nrepresent a homogeneous Euclidean conformation space suitable for\nhigh-dimension optimization and construct funnel-like energy models for\nencounter complexes. Over a protein docking benchmark set and a CAPRI set\nincluding homology docking, we establish that BAL significantly improve against\nboth starting points by rigid docking and refinements by particle swarm\noptimization, providing for one third targets a top-3 near-native prediction.\nBAL also generates tight confidence intervals with half range around 25% of\niRMSD and confidence level at 85%. Its estimated probability of a prediction\nbeing native or not achieves binary classification AUROC at 0.93 and AUPRC over\n0.60 (compared to 0.14 by chance); and also found to help ranking predictions.\nTo the best of our knowledge, this study represents the first uncertainty\nquantification solution for protein docking, with theoretical rigor and\ncomprehensive assessment.\n  Source codes are available at https://github.com/Shen-Lab/BAL.","url_abs":"http://arxiv.org/abs/1902.00067v1","url_pdf":"http://arxiv.org/pdf/1902.00067v1.pdf","source":{"archive":"pwc-archive (Hugging Face), CC BY-SA 4.0","snapshot":"2025-07-28","licence_url":"https://creativecommons.org/licenses/by-sa/4.0/legalcode","row_kind":"abstracts"},"code_links":[{"paper_slug":"bayesian-active-learning-for-optimization-and","repo_url":"https://github.com/Shen-Lab/BAL","is_official":1,"mentioned_in_paper":1,"mentioned_in_github":0,"framework":"none","reach":null}],"tasks":[{"task_slug":"active-learning","task_name":"Active Learning"},{"task_slug":"binary-classification","task_name":"Binary Classification"},{"task_slug":"uncertainty-quantification","task_name":"Uncertainty Quantification"}],"methods":[],"datasets_introduced":[],"methods_introduced":[],"results":[],"syntology":{"atlas_url":null,"mcp":null,"developers":"https://syntology.ai/developers"},"arxiv_metadata":null,"syntology_extracted_results":null}