[{"data":1,"prerenderedAt":-1},["ShallowReactive",2],{"doc-detail-117386-en":3,"doc-seo-117386-105":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":4,"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},117386,7971461740886,"Theodore","https://ap-avatar.wpscdn.com/davatar_3d24733baf745e90a7e4bdd5f77d97b2",8,"Research & Report","Machine Learning Methods for Protein Engineering - Dissertation","This dissertation advances machine learning methods for protein engineering across three key areas. It develops a principled framework that incorporates biophysical knowledge into Bayesian Neural Networks to improve prediction of protein properties, especially in regions far from training data. It also introduces a generative modeling approach enabling controlled protein sequence generation via exact guidance equations for discrete state-space diffusion and flow models. Finally, it presents strategies for designing massive combinatorial protein sequence libraries that can be efficiently assembled and validated experimentally.","UC Berkeley  \nUC Berkeley Electronic Theses and Dissertations  \nTitle  \nMachine Learning Methods for Protein Engineering  \nPermalink  \n[https://escholarship.org/uc/item/16r9155f](https://escholarship.org/uc/item/16r9155f)  \nAuthor  \nNisonoff, Hunter Morris  \nPublication Date  \n2024  \nPeer reviewed|Thesis/dissertation  \n[eScholarship.org](eScholarship.org) Powered by the California Digital Library  \nUniversity of California  \nMachine Learning Methods for Protein Engineering  \nby  \nHunter Nisonoff  \nA dissertation submitted in partial satisfaction of the requirements for the degree of Doctor of Philosophy in Computational Biology in the Graduate Division  \nof the  \nUniversity of California, Berkeley  \nCommittee in charge:  \nProfessor Jennifer Listgarten, Chair Professor Yun S. Song Professor Ian Holmes Professor David Savage  \nFall 2024  \nMachine Learning Methods for Protein Engineering  \nCopyright 2024  \nby  \nHunter Nisonoff  \n1  \nAbstract  \nMachine Learning Methods for Protein Engineering  \nby  \nHunter Nisonoff  \nDoctor of Philosophy in Computational Biology  \nUniversity of California, Berkeley  \nProfessor Jennifer Listgarten, Chair  \nThis dissertation advances machine learning methods for protein engineering across three key areas. First, we develop a principled framework for incorporating biophysical knowledge into Bayesian Neural Networks to enable more robust prediction of protein properties, particularly in regions far from training data. Second, we introduce a new generative modeling approach that enables controlled protein sequence generation by deriving exact guidance equations for discrete state-space diffusion and flow models. Finally, we present strategies for designing massive combinatorial libraries of protein sequences that can be efficiently assembled and tested experimentally. Together, these advances provide a comprehensive toolkit for machine learning-driven protein engineering that bridges computational design and experimental validation.  \ni  \nTo my parents, Jennifer and Philip Nisonoff  \nii  \nContents  \nContents ii  \nList of Figures iv  \nList of Tables v  \n1 Introduction 1  \n2 Blending Biophysics with Neural Networks 6  \n2.1 Introduction .................................... 6  \n2.2 Results ....................................... 10  \n2.3 Discussion ..................................... 17  \n2.4 Methods ...................................... 18  \n3 Guidance for Discrete Diffusion and Flow Models 22  \n3.1 Introduction .................................... 22  \n3.2 Results ....................................... 27  \n3.3 Discussion ..................................... 36  \n3.4 Methods ...................................... 36  \n4 Large-Scale Design of Protein Families 47  \n4.1 Introduction .................................... 47  \n4.2 Results ....................................... 49  \n4.3 Discussion ..................................... 54  \n4.4 Methods ...................................... 55  \n5 Conclusion 64  \nBibliography 66  \nA Supplementary Material for Chapter 2 84  \nA.1 Derivation and additional details of f v-BNN .................. 84  \nA.2 Incorporating Full Covariance Matrices ..................... 87  \niii  \nA.3 Applicability to Protein Engineering ....................... 88  \nA.4 Protein Fitness Tasks with Evolutionary Information ............. 90  \nA.5 Statistical Significance .............................. 93  \nB Supplementary Material for Chapter 3 96  \nB.1 Implementation Details .............................. 96  \nB.2 Discussion of Related Work ........................... 105  \nB.3 Experimental Details ............................... 109  \niv  \nList of Figures  \n1.1 Design, Build, Test, Learn (DBTL) cycle ...................... 4  \n2.1 Illustration of function-value-prior-augmented BNNs ............... 8  \n3.1 Overview of discrete state-space guidance ...................... 24  \n3.2 Property-guided molecule generation ........................ 31  \n3.3 Cell type-specific DNA enhancer genera","cbCaiiNQ58je7IXe","https://ap.wps.com/l/cbCaiiNQ58je7IXe","pdf",11843999,1,148,"English","en",105,"# Introduction\n# Blending Biophysics with Neural Networks\n## Results\n## Methods\n# Guidance for Discrete Diffusion and Flow Models\n## Results\n## Methods\n# Large-Scale Design of Protein Families\n## Results\n## Methods\n# Conclusion\n# Supplementary Material for Chapter 2\n# Supplementary Material for Chapter 3","[{\"question\":\"How does the dissertation incorporate biophysical knowledge into machine learning models for proteins?\",\"answer\":\"It proposes a framework that blends biophysical knowledge into Bayesian Neural Networks to enhance robustness of protein property prediction, particularly away from training regions.\"},{\"question\":\"What is the generative modeling contribution for protein sequence design?\",\"answer\":\"It introduces a generative modeling approach that enables controlled protein sequence generation by deriving exact guidance equations for discrete state-space diffusion and flow models.\"},{\"question\":\"How does the dissertation address experimental validation of large protein sequence libraries?\",\"answer\":\"It presents strategies for designing massive combinatorial libraries of protein sequences and emphasizes efficient assembly and experimental testing to validate designs.\"}]","Machine Learning Methods for Protein Engineering - Dissertation | PDF",1785675523,373,{"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},"machine-learning-methods-for-protein-engineering-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/machine-learning-methods-for-protein-engineering-dissertation/117386/",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","2026-08-02",true,{"@type":65,"interactionType":66,"userInteractionCount":4},"InteractionCounter",{"@type":67},"ViewAction",{"@type":69,"mainEntity":70},"FAQPage",[71,77,81],{"name":72,"@type":73,"acceptedAnswer":74},"How does the dissertation incorporate biophysical knowledge into machine learning models for proteins?","Question",{"text":75,"@type":76},"It proposes a framework that blends biophysical knowledge into Bayesian Neural Networks to enhance robustness of protein property prediction, particularly away from training regions.","Answer",{"name":78,"@type":73,"acceptedAnswer":79},"What is the generative modeling contribution for protein sequence design?",{"text":80,"@type":76},"It introduces a generative modeling approach that enables controlled protein sequence generation by deriving exact guidance equations for discrete state-space diffusion and flow models.",{"name":82,"@type":73,"acceptedAnswer":83},"How does the dissertation address experimental validation of large protein sequence libraries?",{"text":84,"@type":76},"It presents strategies for designing massive combinatorial libraries of protein sequences and emphasizes efficient assembly and experimental testing to validate designs.","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"]