[{"data":1,"prerenderedAt":-1},["ShallowReactive",2],{"doc-detail-117830-en":3,"doc-seo-117830-105":30,"detail-sidebar-cat-0-en-105":96},{"code":4,"msg":5,"data":6},0,"success",{"doc_id":7,"user_id":8,"nickname":9,"user_avatar":10,"doc_module":4,"category_id":11,"category_name":12,"doc_title":13,"doc_description":14,"doc_content":15,"file_id":16,"file_url":17,"file_type":18,"file_size":19,"view_count":20,"is_deleted":4,"is_public":20,"is_downloadable":20,"audit_status":20,"page_count":21,"language":22,"language_code":23,"site_id":24,"html_lang":23,"table_of_contents":25,"faqs":26,"seo_title":27,"seo_description":14,"update_tm":28,"read_time":29},117830,549758146520,"Patrick","https://ap-avatar.wpscdn.com/avatar/80002397d8c0411e94?_k=1775819394049821470",8,"Research & Report","Deep Learning Applications in Structure-Based Drug Discovery - Doctoral Thesis","Recent advances in algorithms and computing hardware have accelerated machine learning and deep learning across scientific fields. A key research direction is deep learning for structure-based drug design, aiming to generate effective drugs against a chosen pharmacological target. This thesis investigates deep learning in early drug discovery, focusing on structure-based virtual screening, binding affinity prediction, and de novo drug design. Docking with flexible residues is implemented in Gnina, followed by improvements to scoring functions and evaluation of a generative model for industrial pipelines.","Deep Learning Applications in Structure-Based Drug Discovery  \nRocco Meli  \nLinacre College University of Oxford  \nA thesis submitted for the degree of Doctor of Philosophy  \nOctober 1, 2022  \nThis page intentionally contains only this sentence.  \n© Rocco Meli, 2019-2022  \nThis work is licensed under the Creative Commons Attribution-ShareAlike 4.0 International License (CC BY-SA 4.0) . To view a copy of this license, visit [http://creativecommons.org/licenses/by-sa/4.0/](http://creativecommons.org/licenses/by-sa/4.0/ or)[ or](http://creativecommons.org/licenses/by-sa/4.0/ or) send a letter to Creative Commons, PO Box 1866, Mountain View, CA 94042, USA.  \nThis page intentionally contains only this sentence.  \nTo Laura & Edy.  \nTo Pietro.  \nTo Marta.  \nThis page intentionally contains only this sentence.  \nAbstract  \nIn recent years, machine learning and deep learning applications have permeated all fields of science thanks to rapid algorithmic advances and computer hardware developments. A very active area of research is the use of deep learning in structure-based drug design, where the goal is to design effective drugs against a pharmacological target of interest.  \nIn this work, we explored the use of deep learning in the early stages of drug discovery. In particular, we focussed on structure-based virtual screening, binding affinity prediction, and de novo drug design.  \nFirst, we enabled docking with flexible residues within Gnina, a state-of-theart docking software based on convolutional neural networks, and we performed a large-scale cross-docking study of such methodology, outlining its strengthsand weaknesses.  \nSecond, we extracted the convolutional neural network scoring function from the docking software into a standalone package for fast prototyping. With the new software at hand, we explored different annotations for supervised learning to improve the convolutional neural network scoring function for docking with flexible residues.  \nThird, we developed a novel scoring function for binding affinity prediction based on a successful deep learning architecture used to develop machine learning force fields.  \nFinally, we carefully evaluate a generative model for de novo design for the application in industrial drug discovery pipelines. We outline the weaknesses of the method and the problems with current evaluations of generative models.  \nThis page intentionally contains only this sentence.  \nAcknowledgments  \nThe path toward a DPhil is a complex and difficult one, especially when it brings you far from home, and to work in yet another language that is not yours. If ontop of the expected and exciting challenges of trying to do innovative research you add a global pandemic, things get very tricky. More than 50% of the research described in this work has been performed in the (scientific) solitude of my old room at my parent’s house or the living room at my partner’s house. Therefore, I have a lot of people to thank for all the help and support they provided over the years.  \nPhilip Biggin—my main supervisor—for all the guidance and support provided during my DPhil, and for fostering a friendly and relaxed research group. I particularly appreciated the freedom I was given and his constant availability. Despite the large size of the research group, he has always been happy to discuss my progress and my blockers. His support has been instrumental during the lockdowns and allowed me to build a decent work environment at home.  \nGarrett Morris—my second supervisor—for additional guidance, suggestions, and the “pedantic” corrections (as he calls them himself), which greatly improved the wording of some manuscripts associated with this work.  \nEvotec for providing additional funding—particularly helpful during the SARS-CoV-2 pandemic, and the following cost of living crisis—, and its scientists for providing invaluable help and comments on the different projects.  \nAndrew Anighoro for closely following my DPhil progress durin","cbCaikLZC7rl8Ayd","https://ap.wps.com/l/cbCaikLZC7rl8Ayd","pdf",42369737,1,362,"English","en",105,"# Abstract\n## Docking with flexible residues\n## CNN scoring function extraction and supervised learning\n## Novel binding affinity scoring function\n## Generative modeling for de novo design\n## Acknowledgments","[{\"question\":\"What areas of structure-based drug discovery does the thesis focus on?\",\"answer\":\"It focuses on structure-based virtual screening, binding affinity prediction, and de novo drug design in the early stages of discovery.\"},{\"question\":\"How is Gnina used in the work?\",\"answer\":\"The thesis enables docking with flexible residues within Gnina and performs a large-scale cross-docking study to evaluate its strengths and weaknesses.\"},{\"question\":\"What contributions are made to improve docking scoring functions?\",\"answer\":\"It extracts the convolutional neural network scoring function from Gnina into a standalone package, then explores different supervised-learning annotations to improve scoring for flexible-residue docking.\"},{\"question\":\"How is generative modeling evaluated for de novo design?\",\"answer\":\"The work carefully evaluates a generative model for de novo design, outlining weaknesses of the method and identifying problems with current evaluation practices for generative models.\"}]","Deep Learning Applications in Structure-Based Drug Discovery - Doctoral Thesis | PDF",1785679865,912,{"code":4,"msg":31,"data":32},"ok",{"site_id":24,"language":23,"slug":33,"title":13,"keywords":34,"description":14,"schema_data":35,"social_meta":91,"head_meta":93,"extra_data":95,"updated_unix":28},"deep-learning-applications-in-structure-based-drug-discovery-doctoral-thesis","",{"@graph":36,"@context":90},[37,54,69],{"@type":38,"itemListElement":39},"BreadcrumbList",[40,44,48,51],{"item":41,"name":42,"@type":43,"position":20},"https://docshare.wps.com","Home","ListItem",{"item":45,"name":46,"@type":43,"position":47},"https://docshare.wps.com/document/","Document",2,{"item":49,"name":12,"@type":43,"position":50},"https://docshare.wps.com/document/research-report/",3,{"item":52,"name":13,"@type":43,"position":53},"https://docshare.wps.com/document/deep-learning-applications-in-structure-based-drug-discovery-doctoral-thesis/117830/",4,{"url":52,"name":13,"@type":55,"author":56,"headline":13,"publisher":58,"fileFormat":61,"inLanguage":23,"description":14,"dateModified":62,"datePublished":63,"encodingFormat":61,"isAccessibleForFree":64,"interactionStatistic":65},"DigitalDocument",{"name":9,"@type":57},"Person",{"url":41,"name":59,"@type":60},"DocShare","Organization","application/pdf","2026-08-05","2026-08-02",true,{"@type":66,"interactionType":67,"userInteractionCount":20},"InteractionCounter",{"@type":68},"ViewAction",{"@type":70,"mainEntity":71},"FAQPage",[72,78,82,86],{"name":73,"@type":74,"acceptedAnswer":75},"What areas of structure-based drug discovery does the thesis focus on?","Question",{"text":76,"@type":77},"It focuses on structure-based virtual screening, binding affinity prediction, and de novo drug design in the early stages of discovery.","Answer",{"name":79,"@type":74,"acceptedAnswer":80},"How is Gnina used in the work?",{"text":81,"@type":77},"The thesis enables docking with flexible residues within Gnina and performs a large-scale cross-docking study to evaluate its strengths and weaknesses.",{"name":83,"@type":74,"acceptedAnswer":84},"What contributions are made to improve docking scoring functions?",{"text":85,"@type":77},"It extracts the convolutional neural network scoring function from Gnina into a standalone package, then explores different supervised-learning annotations to improve scoring for flexible-residue docking.",{"name":87,"@type":74,"acceptedAnswer":88},"How is generative modeling evaluated for de novo design?",{"text":89,"@type":77},"The work carefully evaluates a generative model for de novo design, outlining weaknesses of the method and identifying problems with current evaluation practices for generative models.","https://schema.org",{"og:url":52,"og:type":92,"og:title":13,"og:site_name":59,"og:description":14},"article",{"robots":94,"canonical":52},"index,follow",{"doc_id":7,"site_id":24},{"code":4,"msg":5,"data":97},[98,102,106,110,115,120,125,128,133,136,140],{"id":20,"doc_module":4,"doc_module_name":46,"category_name":99,"show_sort_weight":100,"slug":101},"Story & Novel",90,"story-novel",{"id":47,"doc_module":4,"doc_module_name":46,"category_name":103,"show_sort_weight":104,"slug":105},"Literature",80,"literature",{"id":53,"doc_module":4,"doc_module_name":46,"category_name":107,"show_sort_weight":108,"slug":109},"Exam",70,"exam",{"id":111,"doc_module":4,"doc_module_name":46,"category_name":112,"show_sort_weight":113,"slug":114},5,"Comic",60,"comic",{"id":116,"doc_module":4,"doc_module_name":46,"category_name":117,"show_sort_weight":118,"slug":119},6,"Technology",50,"technology",{"id":121,"doc_module":4,"doc_module_name":46,"category_name":122,"show_sort_weight":123,"slug":124},7,"Healthcare",40,"healthcare",{"id":11,"doc_module":4,"doc_module_name":46,"category_name":12,"show_sort_weight":126,"slug":127},30,"research-report",{"id":129,"doc_module":4,"doc_module_name":46,"category_name":130,"show_sort_weight":131,"slug":132},9,"Religion & Spirituality",20,"religion-spirituality",{"id":131,"doc_module":4,"doc_module_name":46,"category_name":134,"show_sort_weight":131,"slug":135},"World Cup","world-cup",{"id":137,"doc_module":4,"doc_module_name":46,"category_name":138,"show_sort_weight":137,"slug":139},10,"Lifestyle","lifestyle",{"id":141,"doc_module":4,"doc_module_name":46,"category_name":142,"show_sort_weight":111,"slug":143},19,"General","general"]