[{"data":1,"prerenderedAt":-1},["ShallowReactive",2],{"doc-detail-123433-en":3,"doc-seo-123433-105":30,"detail-sidebar-cat-0-en-105":95},{"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},123433,1374391974564,"Clementine","https://ap-avatar.wpscdn.com/avatar/14000253aa45c000a9e?x-image-process=image/resize,m_fixed,w_180,h_180&k=1779874745381141002",8,"Research & Report","Machine Learning Approaches to Understanding Codon Choice - Doctor of Philosophy Dissertation","Identical proteins can be encoded by different synonymous codons, which are translated by ribosomes at different rates. The thesis addresses how codon choice shapes biological processes, linking codon-driven translation dynamics to disorders from synonymous mutations and to the design of synthetic mRNAs. Using protein language models, it introduces a machine learning method to predict codon choice from amino acid sequence, incorporating position and protein structure constraints in yeast. In parallel, it performs a genome-wide Cas9 retron editing screen to map fitness effects of thousands of synonymous substitutions. Finally, it develops Trias, a generative codon-language model that produces realistic mRNA sequences with high protein output for human sequences.","UC Berkeley  \nUC Berkeley Electronic Theses and Dissertations  \nTitle  \nMachine Learning Approaches to Understanding Codon Choice  \nPermalink  \n[https://escholarship.org/uc/item/1f49173f](https://escholarship.org/uc/item/1f49173f)  \nISBN  \n9798297601031  \nAuthor  \nSakharova, Helen Alexandra  \nPublication Date  \n2025-08-01  \nPeer reviewed|Thesis/dissertation  \n[eScholarship.org](eScholarship.org) Powered by the California Digital Library  \nUniversity of California  \nMachine Learning Approaches to Understanding Codon Choice  \nBy  \nHelen Alexandra Sakharova  \nA dissertation submitted in partial satisfaction of the requirements for the degree of Doctor of Philosophy  \nin Computational Biology in the  \nGraduate Division  \nof the  \nUniversity of California, Berkeley  \nCommittee in charge:  \nAssistant Professor Liana Lareau, Chair Professor Jamie Cate  \nProfessor Ian Holmes Associate Professor Nicholas Ingolia  \nSummer 2025  \nMachine Learning Approaches to Understanding Codon Choice  \nCopyright 2025  \nby  \nHelen Alexandra Sakharova  \n1  \nAbstract  \nMachine Learning Approaches to Understanding Codon Choice  \nby  \nHelen Alexandra Sakharova  \nDoctor of Philosophy in Computational Biology  \nUniversity of California, Berkeley  \nAssistant Professor Liana Lareau, Chair  \nIdentical proteins can be encoded in DNA using different synonymous codons, which are translated by the ribosome at different rates. The mechanisms by and extent to which codon choice impacts biological processes remains a fundamental open question. Elucidating the rules governing codon choice is vital both to understanding disorders caused by synonymous mutations, and to improve our ability to design synthetic mRNAs. The structure and function of a protein may set requirements on the process of translation that create pressure to select for slower or faster translated codons. Leveraging existing protein language models, I build a machine learning model to predict codon choice from amino acid sequence. My model effectively combines information about position and protein structure to learn subtle but wide-reaching constraints on codon choice in yeast. In parallel, I conduct a genome-widescreen in yeast to reliably identify synonymous variants that significantly decrease or increase fitness, using Cas9 retron editing to create thousands of synonymous codon substitutions in endogenous loci. Lastly, we extend our exploration of codon usage to create Trias, a generative codon-language model applicable to human sequences. We demonstrate that Trias can be used to generate realistic mRNA sequences with high protein output.  \ni  \nTo my grandparents.  \nii  \nContents  \nContents ii  \nList of Figures iii  \nList of Tables v  \n1 Introduction 1  \nReferences ........................................ 3  \n2 Constraints on codon choice 6  \n2.1 Introduction .................................... 7  \n2.2 Results ....................................... 9  \n2.3 Discussion ..................................... 19  \n2.4 Methods ...................................... 23  \nReferences ........................................ 29  \n2.5 Supplement .................................... 34  \n3 A generative language model for mRNA design 40  \n3.1 Introduction .................................... 40  \n3.2 Results ....................................... 43  \n3.3 Discussion ..................................... 51  \n3.4 Methods ...................................... 53  \nReferences ........................................ 57  \n3.5 Supplement .................................... 61  \niii  \nList of Figures  \n2.1 Codon choice affects protein production and function................ 8  \n2.2 Model predicts codon choice solely from amino acid sequence........... 11  \n2.3 Model learns generalizable constraints on codon choice from position and protein structures........................................ 13  \n2.4 Genome-wide screen reliably identifies advantageous and deleterious synonymous mutations..........................","cbCaiiJhuuifntmi","https://ap.wps.com/l/cbCaiiJhuuifntmi","pdf",15967470,1,81,"English","en",105,"# Contents\n## Introduction\n## Constraints on codon choice\n## A generative language model for mRNA design\n## References\n## Supplement","[{\"question\":\"Why does codon choice matter even when proteins are identical?\",\"answer\":\"Synonymous codons are translated at different rates, so codon choice can influence translation and downstream biological processes despite producing the same amino acid sequence.\"},{\"question\":\"How does the thesis predict codon choice from biological sequence information?\",\"answer\":\"It builds a machine learning model that leverages protein language models and combines information from position and protein structure to learn constraints on codon choice in yeast.\"},{\"question\":\"What genome-wide method is used to test synonymous variants?\",\"answer\":\"A genome-widescreen in yeast uses Cas9 retron editing to introduce thousands of synonymous codon substitutions at endogenous loci and measure fitness effects.\"},{\"question\":\"What is Trias and what does it generate?\",\"answer\":\"Trias is a generative codon-language model that generates realistic mRNA sequences for human inputs and supports high protein output.\"}]","Machine Learning Approaches to Understanding Codon Choice - Doctor of Philosophy Dissertation | PDF",1785816447,204,{"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":90,"head_meta":92,"extra_data":94,"updated_unix":28},"machine-learning-approaches-to-understanding-codon-choice-doctor-of-philosophy-dissertation","",{"@graph":36,"@context":89},[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-approaches-to-understanding-codon-choice-doctor-of-philosophy-dissertation/123433/",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","2026-08-04",true,{"@type":65,"interactionType":66,"userInteractionCount":4},"InteractionCounter",{"@type":67},"ViewAction",{"@type":69,"mainEntity":70},"FAQPage",[71,77,81,85],{"name":72,"@type":73,"acceptedAnswer":74},"Why does codon choice matter even when proteins are identical?","Question",{"text":75,"@type":76},"Synonymous codons are translated at different rates, so codon choice can influence translation and downstream biological processes despite producing the same amino acid sequence.","Answer",{"name":78,"@type":73,"acceptedAnswer":79},"How does the thesis predict codon choice from biological sequence information?",{"text":80,"@type":76},"It builds a machine learning model that leverages protein language models and combines information from position and protein structure to learn constraints on codon choice in yeast.",{"name":82,"@type":73,"acceptedAnswer":83},"What genome-wide method is used to test synonymous variants?",{"text":84,"@type":76},"A genome-widescreen in yeast uses Cas9 retron editing to introduce thousands of synonymous codon substitutions at endogenous loci and measure fitness effects.",{"name":86,"@type":73,"acceptedAnswer":87},"What is Trias and what does it generate?",{"text":88,"@type":76},"Trias is a generative codon-language model that generates realistic mRNA sequences for human inputs and supports high protein output.","https://schema.org",{"og:url":52,"og:type":91,"og:title":13,"og:site_name":59,"og:description":14},"article",{"robots":93,"canonical":52},"index,follow",{"doc_id":7,"site_id":24},{"code":4,"msg":5,"data":96},[97,101,105,109,114,119,124,127,132,135,139],{"id":20,"doc_module":4,"doc_module_name":46,"category_name":98,"show_sort_weight":99,"slug":100},"Story & Novel",90,"story-novel",{"id":47,"doc_module":4,"doc_module_name":46,"category_name":102,"show_sort_weight":103,"slug":104},"Literature",80,"literature",{"id":53,"doc_module":4,"doc_module_name":46,"category_name":106,"show_sort_weight":107,"slug":108},"Exam",70,"exam",{"id":110,"doc_module":4,"doc_module_name":46,"category_name":111,"show_sort_weight":112,"slug":113},5,"Comic",60,"comic",{"id":115,"doc_module":4,"doc_module_name":46,"category_name":116,"show_sort_weight":117,"slug":118},6,"Technology",50,"technology",{"id":120,"doc_module":4,"doc_module_name":46,"category_name":121,"show_sort_weight":122,"slug":123},7,"Healthcare",40,"healthcare",{"id":11,"doc_module":4,"doc_module_name":46,"category_name":12,"show_sort_weight":125,"slug":126},30,"research-report",{"id":128,"doc_module":4,"doc_module_name":46,"category_name":129,"show_sort_weight":130,"slug":131},9,"Religion & Spirituality",20,"religion-spirituality",{"id":130,"doc_module":4,"doc_module_name":46,"category_name":133,"show_sort_weight":130,"slug":134},"World Cup","world-cup",{"id":136,"doc_module":4,"doc_module_name":46,"category_name":137,"show_sort_weight":136,"slug":138},10,"Lifestyle","lifestyle",{"id":140,"doc_module":4,"doc_module_name":46,"category_name":141,"show_sort_weight":110,"slug":142},19,"General","general"]