{"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/predicting-protein-inter-residue-contacts","title":"Predicting protein inter-residue contacts using composite likelihood maximization and deep learning","arxiv_id":"1809.00083","date":"2018-08-31","proceeding":null,"authors":["Haicang Zhang","Qi Zhang","Fusong Ju","Jianwei Zhu","Yujuan Gao","Ziwei Xie","Minghua Deng","Shiwei Sun","Wei-Mou Zheng","Dongbo Bu"],"abstract":"Accurate prediction of inter-residue contacts of a protein is important to\ncalcu- lating its tertiary structure. Analysis of co-evolutionary events among\nresidues has been proved effective to inferring inter-residue contacts. The\nMarkov ran- dom field (MRF) technique, although being widely used for contact\nprediction, suffers from the following dilemma: the actual likelihood function\nof MRF is accurate but time-consuming to calculate, in contrast, approximations\nto the actual likelihood, say pseudo-likelihood, are efficient to calculate but\ninaccu- rate. Thus, how to achieve both accuracy and efficiency simultaneously\nremains a challenge. In this study, we present such an approach (called clmDCA)\nfor contact prediction. Unlike plmDCA using pseudo-likelihood, i.e., the\nproduct of conditional probability of individual residues, our approach uses\ncomposite- likelihood, i.e., the product of conditional probability of all\nresidue pairs. Com- posite likelihood has been theoretically proved as a better\napproximation to the actual likelihood function than pseudo-likelihood.\nMeanwhile, composite likelihood is still efficient to maximize, thus ensuring\nthe efficiency of clmDCA. We present comprehensive experiments on popular\nbenchmark datasets, includ- ing PSICOV dataset and CASP-11 dataset, to show\nthat: i) clmDCA alone outperforms the existing MRF-based approaches in\nprediction accuracy. ii) When equipped with deep learning technique for\nrefinement, the prediction ac- curacy of clmDCA was further significantly\nimproved, suggesting the suitability of clmDCA for subsequent refinement\nprocedure. We further present successful application of the predicted contacts\nto accurately build tertiary structures for proteins in the PSICOV dataset.\n  Accessibility: The software clmDCA and a server are publicly accessible\nthrough http://protein.ict.ac.cn/clmDCA/.","url_abs":"http://arxiv.org/abs/1809.00083v1","url_pdf":"http://arxiv.org/pdf/1809.00083v1.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":"predicting-protein-inter-residue-contacts","repo_url":"https://github.com/zhanghaicang/MRF-suite","is_official":0,"mentioned_in_paper":0,"mentioned_in_github":1,"framework":"none","reach":null}],"tasks":[{"task_slug":"prediction","task_name":"Prediction"}],"methods":[],"datasets_introduced":[],"methods_introduced":[],"results":[],"syntology":{"atlas_url":null,"mcp":null,"developers":"https://syntology.ai/developers"},"arxiv_metadata":null,"syntology_extracted_results":null}