[{"data":1,"prerenderedAt":-1},["ShallowReactive",2],{"doc-detail-121434-en":3,"doc-seo-121434-105":30,"detail-sidebar-cat-0-en-105":91},{"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},121434,962075114101,"Seraphina","https://ap-avatar.wpscdn.com/avatar/e000253a75eb197efd?x-image-process=image/resize,m_fixed,w_180,h_180&k=1780044092746381165",8,"Research & Report","Explainable Machine Learning and Applications in Protein-Ligand Complex Structure Prediction - Doctor of Philosophy Dissertation","Explainable machine learning methods are developed to interpret model behavior in protein-focused settings, with per-instance feature attribution grounded in local linear approximations. Multiple techniques are proposed, including expected gradients, attribution priors, and integrated hessians, extending interpretability to feature interactions and more training-stable, interpretable models. A second focus trains protein language models and designs semi-supervised pretraining tasks using ideas from multiple sequence alignments, including profile prediction and seq2msa. Finally, protein structure prediction is coupled with small-molecule docking via a structure prediction network, RoseTTAFold All-Atom, enabling applications in binding-structure inference and virtual screening.","©Copyright 2024  \nPascal Sturmfels  \nExplainable Machine Learning and Applications in Protein-Ligand  \nComplex Structure Prediction  \nPascal Sturmfels  \nA dissertation  \nsubmitted in partial fulfillment of the  \nrequirements for the degree of  \nDoctor of Philosophy  \nUniversity of Washington  \n2024  \nReading Committee:  \nDavid Baker, Chair  \nSheng Wang  \nFrank Dimaio  \nProgram Authorized to Offer Degree:  \nComputer Science & Engineering  \nUniversity of Washington  \nAbstract  \nExplainable Machine Learning and Applications in Protein-Ligand Complex Structure Prediction  \nPascal Sturmfels  \nChair of the Supervisory Committee:  \nDavid Baker  \nBiochemistry  \nThis thesis touches upon two main topics: interpreting machine learning models, and the application of machine learning to protein sequences and structures. The first portion deals with feature attribution techniques, which attribute attribution scores on a per-instance basis to a machine learning model that represent that model locally around that instance as linear. Three methods are proposed: expected gradients, attribution priors, and integrated hessians, that extend interpretability beyond feature attribution towards feature interaction and training more interpretable models. The second portion deals with training protein language models and how best to design semi-supervised pre-training tasks. It takes inspiration from multiple sequence alignments to propose two tasksprofile prediction and seq2msa-that extend language modeling beyond autoregressive and masked language modeling. The third portion deals applications in protein structure prediction, chiefly with predicting the structure of proteins in concert with small molecules. Jointly determining both the structure of a protein from its sequence input and how small molecule binding partners dock to that structure remains an open and challenging problem, and has applications in biological discovery, virtual screening, and de-novo design. This thesis discusses the development of a structure prediction network, RoseTTAFold All-Atom, capable of simultaneous folding and docking, as well as some applications enabled by that existing network.  \nTABLE OF CONTENTS  \nPage  \nList of Figures .......................................... iii  \nList of Tables .......................................... viii  \nChapter 1: Introduction ................................... 1  \n1.1 Explainable Machine Learning ............................ 1  \n1.2 Protein Language Modeling ............................. 2  \n1.3 Protein-Ligand Co-folding .............................. 3  \n1.4 Summary ....................................... 4  \nChapter 2: Explainable Machine Learning ......................... 6  \n2.1 Feature Attribution .................................. 6  \n2.2 Game Theoretic Explanations ............................. 6  \n2.3 On Attribution Baselines in Game-Theoretic Explanations ............. 8  \n2.4 Axiomatic Feature Interactions ............................ 12  \n2.5 Explaining Explanations with Integrated Hessians .................. 14  \n2.6 Smoothing ReLU Networks ............................. 20  \n2.7 Attribution Priors ................................... 23  \n2.8 Discussion ....................................... 27  \nChapter 3: Protein Language Modeling .......................... 31  \n3.1 Motivation ....................................... 31  \n3.2 Masked Language Modeling ............................. 32  \n3.3 Profile Prediction ................................... 32  \n3.4 Results ......................................... 37  \n3.5 Seq2MSA ....................................... 40  \n3.6 Discussion ....................................... 52  \nChapter 4: Protein-Ligand Co-folding ........................... 55  \n4.1 Motivation ....................................... 55  \n4.2 The RoseTTAFold All-Atom Architecture ...................... 56  \n4.3 Training Regime ................................... 59  \n4.4 Ligand Docking ......................","cbCaicqptQ6mLL4m","https://ap.wps.com/l/cbCaicqptQ6mLL4m","pdf",6164847,1,116,"English","en",105,"# Abstract\n# Chapter 1: Introduction\n## Explainable Machine Learning\n## Protein Language Modeling\n## Protein-Ligand Co-folding\n# Chapter 2: Explainable Machine Learning\n## Feature Attribution\n## Game Theoretic Explanations\n## On Attribution Baselines in Game-Theoretic Explanations\n## Axiomatic Feature Interactions\n## Explaining Explanations with Integrated Hessians\n## Smoothing ReLU Networks\n## Attribution Priors\n## Discussion\n# Chapter 3: Protein Language Modeling\n## Motivation\n## Masked Language Modeling\n## Profile Prediction\n## Results\n## Seq2MSA\n## Discussion\n# Chapter 4: Protein-Ligand Co-folding\n## Motivation\n## The RoseTTAFold All-Atom Architecture\n## Training Regime\n## Ligand Docking\n## Applications in Virtual Screening\n## Improving Docking Performance\n## Discussion\n# Chapter 5: Conclusions","[{\"question\":\"What two main topics does the thesis address?\",\"answer\":\"The thesis addresses interpreting machine learning models and applying machine learning to protein sequences and protein structures, especially in protein-ligand settings.\"},{\"question\":\"How does the thesis extend interpretability beyond feature attribution?\",\"answer\":\"It proposes methods such as expected gradients, attribution priors, and integrated hessians to capture feature interactions and support training more interpretable models.\"},{\"question\":\"What is RoseTTAFold All-Atom and what capability does it provide?\",\"answer\":\"RoseTTAFold All-Atom is a structure prediction network designed to perform simultaneous folding and docking, jointly inferring protein structure from sequence input and small-molecule binding partners.\"}]","Explainable Machine Learning and Applications in Protein-Ligand Complex Structure Prediction - Doctor of Philosophy Dissertation | PDF",1785735645,292,{"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":86,"head_meta":88,"extra_data":90,"updated_unix":28},"explainable-machine-learning-and-applications-in-protein-ligand-complex-structure-prediction-doctor-of-philosophy-dissertation","",{"@graph":36,"@context":85},[37,54,68],{"@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/explainable-machine-learning-and-applications-in-protein-ligand-complex-structure-prediction-doctor-of-philosophy-dissertation/121434/",4,{"url":52,"name":13,"@type":55,"author":56,"headline":13,"publisher":58,"fileFormat":61,"inLanguage":23,"description":14,"dateModified":62,"datePublished":62,"encodingFormat":61,"isAccessibleForFree":63,"interactionStatistic":64},"DigitalDocument",{"name":9,"@type":57},"Person",{"url":41,"name":59,"@type":60},"DocShare","Organization","application/pdf","2026-08-03",true,{"@type":65,"interactionType":66,"userInteractionCount":20},"InteractionCounter",{"@type":67},"ViewAction",{"@type":69,"mainEntity":70},"FAQPage",[71,77,81],{"name":72,"@type":73,"acceptedAnswer":74},"What two main topics does the thesis address?","Question",{"text":75,"@type":76},"The thesis addresses interpreting machine learning models and applying machine learning to protein sequences and protein structures, especially in protein-ligand settings.","Answer",{"name":78,"@type":73,"acceptedAnswer":79},"How does the thesis extend interpretability beyond feature attribution?",{"text":80,"@type":76},"It proposes methods such as expected gradients, attribution priors, and integrated hessians to capture feature interactions and support training more interpretable models.",{"name":82,"@type":73,"acceptedAnswer":83},"What is RoseTTAFold All-Atom and what capability does it provide?",{"text":84,"@type":76},"RoseTTAFold All-Atom is a structure prediction network designed to perform simultaneous folding and docking, jointly inferring protein structure from sequence input and small-molecule binding partners.","https://schema.org",{"og:url":52,"og:type":87,"og:title":13,"og:site_name":59,"og:description":14},"article",{"robots":89,"canonical":52},"index,follow",{"doc_id":7,"site_id":24},{"code":4,"msg":5,"data":92},[93,97,101,105,110,115,120,123,128,131,135],{"id":20,"doc_module":4,"doc_module_name":46,"category_name":94,"show_sort_weight":95,"slug":96},"Story & Novel",90,"story-novel",{"id":47,"doc_module":4,"doc_module_name":46,"category_name":98,"show_sort_weight":99,"slug":100},"Literature",80,"literature",{"id":53,"doc_module":4,"doc_module_name":46,"category_name":102,"show_sort_weight":103,"slug":104},"Exam",70,"exam",{"id":106,"doc_module":4,"doc_module_name":46,"category_name":107,"show_sort_weight":108,"slug":109},5,"Comic",60,"comic",{"id":111,"doc_module":4,"doc_module_name":46,"category_name":112,"show_sort_weight":113,"slug":114},6,"Technology",50,"technology",{"id":116,"doc_module":4,"doc_module_name":46,"category_name":117,"show_sort_weight":118,"slug":119},7,"Healthcare",40,"healthcare",{"id":11,"doc_module":4,"doc_module_name":46,"category_name":12,"show_sort_weight":121,"slug":122},30,"research-report",{"id":124,"doc_module":4,"doc_module_name":46,"category_name":125,"show_sort_weight":126,"slug":127},9,"Religion & Spirituality",20,"religion-spirituality",{"id":126,"doc_module":4,"doc_module_name":46,"category_name":129,"show_sort_weight":126,"slug":130},"World Cup","world-cup",{"id":132,"doc_module":4,"doc_module_name":46,"category_name":133,"show_sort_weight":132,"slug":134},10,"Lifestyle","lifestyle",{"id":136,"doc_module":4,"doc_module_name":46,"category_name":137,"show_sort_weight":106,"slug":138},19,"General","general"]