{"id":"https://openalex.org/W2951203227","doi":"https://doi.org/10.1101/gr.224964.117","title":"Deep learning of the regulatory grammar of yeast 5\u2032 untranslated regions from 500,000 random sequences","display_name":"Deep learning of the regulatory grammar of yeast 5\u2032 untranslated regions from 500,000 random sequences","publication_year":2017,"publication_date":"2017-11-02","ids":{"openalex":"https://openalex.org/W2951203227","doi":"https://doi.org/10.1101/gr.224964.117","mag":"2951203227","pmid":"https://pubmed.ncbi.nlm.nih.gov/29097404","pmcid":"https://www.ncbi.nlm.nih.gov/pmc/articles/5741052"},"language":"en","primary_location":{"id":"doi:10.1101/gr.224964.117","doi":"https://doi.org/10.1101/gr.224964.117","is_oa":false,"landing_page_url":"https://doi.org/10.1101/gr.224964.117","pdf_url":null,"source":{"id":"https://openalex.org/S43092948","display_name":"Genome Research","issn_l":"1088-9051","issn":["1088-9051","1549-5469"],"is_oa":false,"is_in_doaj":false,"is_core":true,"listed_in":["anvur-scientific","cwts-core","dongbi-a","doyens","jufo-3","ki-jl-2","medline","norway-2","poland-200","russia-white-list-1"],"host_organization":"https://openalex.org/P4310315909","host_organization_name":"Cold Spring Harbor Laboratory Press","host_organization_lineage":["https://openalex.org/P4310315909"],"host_organization_lineage_names":["Cold Spring Harbor Laboratory Press"],"type":"journal"},"license":null,"license_id":null,"version":"publishedVersion","is_accepted":true,"is_published":true,"raw_source_name":"Genome Research","raw_type":"journal-article"},"volume":null,"type":"preprint","indexed_in":["crossref","pubmed"],"open_access":{"is_oa":true,"oa_status":"green","oa_url":"https://www.biorxiv.org/content/biorxiv/early/2017/05/19/137547.full.pdf","any_repository_has_fulltext":true},"authorships":[{"author_position":"first","author":{"id":"https://openalex.org/A5073324399","display_name":"Josh T. Cuperus","orcid":"https://orcid.org/0000-0002-8019-7733","observed_orcids":["https://orcid.org/0000-0002-8019-7733"]},"institutions":[{"id":"https://openalex.org/I1344073410","display_name":"Howard Hughes Medical Institute","ror":"https://ror.org/006w34k90","country_code":"US","type":"facility","lineage":["https://openalex.org/I1344073410"]},{"id":"https://openalex.org/I201448701","display_name":"University of Washington","ror":"https://ror.org/00cvxb145","country_code":"US","type":"education","lineage":["https://openalex.org/I201448701"]}],"countries":["US"],"is_corresponding":false,"raw_author_name":"Josh Cuperus","raw_affiliation_strings":["Department of Genome Sciences, University of Washington","Howard Hughes Medical Institute, University of Washington","1Department of Genome Sciences, University of Washington, Seattle, Washington 98195, USA","2Howard Hughes Medical Institute, University of Washington, Seattle, Washington 98195, USA"],"raw_orcid":"https://orcid.org/0000-0002-8019-7733","affiliations":[{"raw_affiliation_string":"Department of Genome Sciences, University of Washington","institution_ids":["https://openalex.org/I201448701"]},{"raw_affiliation_string":"Howard Hughes Medical Institute, University of Washington","institution_ids":["https://openalex.org/I1344073410","https://openalex.org/I201448701"]},{"raw_affiliation_string":"1Department of Genome Sciences, University of Washington, Seattle, Washington 98195, USA","institution_ids":["https://openalex.org/I201448701"]},{"raw_affiliation_string":"2Howard Hughes Medical Institute, University of Washington, Seattle, Washington 98195, USA","institution_ids":["https://openalex.org/I1344073410","https://openalex.org/I201448701"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5011037544","display_name":"Benjamin Groves","orcid":"https://orcid.org/0000-0002-4827-5500","observed_orcids":["https://orcid.org/0000-0002-4827-5500"]},"institutions":[{"id":"https://openalex.org/I201448701","display_name":"University of Washington","ror":"https://ror.org/00cvxb145","country_code":"US","type":"education","lineage":["https://openalex.org/I201448701"]}],"countries":["US"],"is_corresponding":false,"raw_author_name":"Benjamin Groves","raw_affiliation_strings":["Department of Electrical Engineering, University of Washington","3Department of Electrical Engineering, University of Washington, Seattle, Washington 98195, USA"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"Department of Electrical Engineering, University of Washington","institution_ids":["https://openalex.org/I201448701"]},{"raw_affiliation_string":"3Department of Electrical Engineering, University of Washington, Seattle, Washington 98195, USA","institution_ids":["https://openalex.org/I201448701"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5102848527","display_name":"Anna Kuchina","orcid":"https://orcid.org/0000-0002-7267-101X","observed_orcids":["https://orcid.org/0000-0002-7267-101X"]},"institutions":[{"id":"https://openalex.org/I201448701","display_name":"University of Washington","ror":"https://ror.org/00cvxb145","country_code":"US","type":"education","lineage":["https://openalex.org/I201448701"]}],"countries":["US"],"is_corresponding":false,"raw_author_name":"Anna Kuchina","raw_affiliation_strings":["Department of Electrical Engineering, University of Washington","3Department of Electrical Engineering, University of Washington, Seattle, Washington 98195, USA"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"Department of Electrical Engineering, University of Washington","institution_ids":["https://openalex.org/I201448701"]},{"raw_affiliation_string":"3Department of Electrical Engineering, University of Washington, Seattle, Washington 98195, USA","institution_ids":["https://openalex.org/I201448701"]}]},{"author_position":"middle","author":{"id":null,"display_name":"Alexander B. Rosenberg","orcid":null,"observed_orcids":[]},"institutions":[{"id":"https://openalex.org/I201448701","display_name":"University of Washington","ror":"https://ror.org/00cvxb145","country_code":"US","type":"education","lineage":["https://openalex.org/I201448701"]}],"countries":["US"],"is_corresponding":false,"raw_author_name":"Alexander B. Rosenberg","raw_affiliation_strings":["Department of Electrical Engineering, University of Washington","3Department of Electrical Engineering, University of Washington, Seattle, Washington 98195, USA"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"Department of Electrical Engineering, University of Washington","institution_ids":["https://openalex.org/I201448701"]},{"raw_affiliation_string":"3Department of Electrical Engineering, University of Washington, Seattle, Washington 98195, USA","institution_ids":["https://openalex.org/I201448701"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5108373424","display_name":"Neboj\u0161a Joji\u0107","orcid":"https://orcid.org/0009-0009-7115-2267","observed_orcids":["https://orcid.org/0009-0009-7115-2267"]},"institutions":[{"id":"https://openalex.org/I1290206253","display_name":"Microsoft (United States)","ror":"https://ror.org/00d0nc645","country_code":"US","type":"company","lineage":["https://openalex.org/I1290206253"]}],"countries":["US"],"is_corresponding":false,"raw_author_name":"Nebojsa Jojic","raw_affiliation_strings":["Microsoft Research","4Microsoft Research, Seattle, Washington 98195, USA"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"Microsoft Research","institution_ids":["https://openalex.org/I1290206253"]},{"raw_affiliation_string":"4Microsoft Research, Seattle, Washington 98195, USA","institution_ids":["https://openalex.org/I1290206253"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5023041153","display_name":"Stanley Fields","orcid":"https://orcid.org/0000-0001-5504-5925","observed_orcids":["https://orcid.org/0000-0001-5504-5925"]},"institutions":[{"id":"https://openalex.org/I1344073410","display_name":"Howard Hughes Medical Institute","ror":"https://ror.org/006w34k90","country_code":"US","type":"facility","lineage":["https://openalex.org/I1344073410"]},{"id":"https://openalex.org/I201448701","display_name":"University of Washington","ror":"https://ror.org/00cvxb145","country_code":"US","type":"education","lineage":["https://openalex.org/I201448701"]}],"countries":["US"],"is_corresponding":true,"raw_author_name":"Stanley Fields","raw_affiliation_strings":["Department of Genome Sciences, University of Washington","Department of Medicine, University of Washington","Howard Hughes Medical Institute, University of Washington","1Department of Genome Sciences, University of Washington, Seattle, Washington 98195, USA","2Howard Hughes Medical Institute, University of Washington, Seattle, Washington 98195, USA","5Department of Medicine, University of Washington, Seattle, Washington 98195, USA"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"Department of Genome Sciences, University of Washington","institution_ids":["https://openalex.org/I201448701"]},{"raw_affiliation_string":"Department of Medicine, University of Washington","institution_ids":["https://openalex.org/I201448701"]},{"raw_affiliation_string":"Howard Hughes Medical Institute, University of Washington","institution_ids":["https://openalex.org/I1344073410","https://openalex.org/I201448701"]},{"raw_affiliation_string":"1Department of Genome Sciences, University of Washington, Seattle, Washington 98195, USA","institution_ids":["https://openalex.org/I201448701"]},{"raw_affiliation_string":"2Howard Hughes Medical Institute, University of Washington, Seattle, Washington 98195, USA","institution_ids":["https://openalex.org/I1344073410","https://openalex.org/I201448701"]},{"raw_affiliation_string":"5Department of Medicine, University of Washington, Seattle, Washington 98195, USA","institution_ids":["https://openalex.org/I201448701"]}]},{"author_position":"last","author":{"id":"https://openalex.org/A5021731745","display_name":"Georg Seelig","orcid":"https://orcid.org/0000-0002-3163-8782","observed_orcids":["https://orcid.org/0000-0002-3163-8782"]},"institutions":[{"id":"https://openalex.org/I201448701","display_name":"University of Washington","ror":"https://ror.org/00cvxb145","country_code":"US","type":"education","lineage":["https://openalex.org/I201448701"]}],"countries":["US"],"is_corresponding":true,"raw_author_name":"Georg Seelig","raw_affiliation_strings":["Department of Computer Science & Engineering, University of Washington","Department of Electrical Engineering, University of Washington","3Department of Electrical Engineering, University of Washington, Seattle, Washington 98195, USA","6Department of Computer Science & Engineering, University of Washington, Seattle, Washington 98195, USA"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"Department of Computer Science & Engineering, University of Washington","institution_ids":["https://openalex.org/I201448701"]},{"raw_affiliation_string":"Department of Electrical Engineering, University of Washington","institution_ids":["https://openalex.org/I201448701"]},{"raw_affiliation_string":"3Department of Electrical Engineering, University of Washington, Seattle, Washington 98195, USA","institution_ids":["https://openalex.org/I201448701"]},{"raw_affiliation_string":"6Department of Computer Science & Engineering, University of Washington, Seattle, Washington 98195, USA","institution_ids":["https://openalex.org/I201448701"]}]}],"institutions":[],"countries_distinct_count":1,"institutions_distinct_count":3,"corresponding_author_ids":["https://openalex.org/A5021731745","https://openalex.org/A5023041153"],"corresponding_institution_ids":["https://openalex.org/I1344073410","https://openalex.org/I201448701"],"apc_list":null,"apc_paid":null,"fwci":null,"has_fulltext":true,"cited_by_count":205,"citation_normalized_percentile":null,"cited_by_percentile_year":null,"biblio":{"volume":"27","issue":"12","first_page":"2015","last_page":"2024"},"is_retracted":false,"is_paratext":false,"is_xpac":false,"primary_topic":{"id":"https://openalex.org/T10521","display_name":"RNA and protein synthesis mechanisms","score":0.9854000210762024,"subfield":{"id":"https://openalex.org/subfields/1312","display_name":"Molecular Biology"},"field":{"id":"https://openalex.org/fields/13","display_name":"Biochemistry, Genetics and Molecular Biology"},"domain":{"id":"https://openalex.org/domains/1","display_name":"Life Sciences"}},"topics":[{"id":"https://openalex.org/T10521","display_name":"RNA and protein synthesis mechanisms","score":0.9854000210762024,"subfield":{"id":"https://openalex.org/subfields/1312","display_name":"Molecular Biology"},"field":{"id":"https://openalex.org/fields/13","display_name":"Biochemistry, Genetics and Molecular Biology"},"domain":{"id":"https://openalex.org/domains/1","display_name":"Life Sciences"}},{"id":"https://openalex.org/T10604","display_name":"RNA Research and Splicing","score":0.00279999990016222,"subfield":{"id":"https://openalex.org/subfields/1312","display_name":"Molecular Biology"},"field":{"id":"https://openalex.org/fields/13","display_name":"Biochemistry, Genetics and Molecular Biology"},"domain":{"id":"https://openalex.org/domains/1","display_name":"Life Sciences"}},{"id":"https://openalex.org/T10497","display_name":"Fungal and yeast genetics research","score":0.002199999988079071,"subfield":{"id":"https://openalex.org/subfields/1312","display_name":"Molecular Biology"},"field":{"id":"https://openalex.org/fields/13","display_name":"Biochemistry, Genetics and Molecular Biology"},"domain":{"id":"https://openalex.org/domains/1","display_name":"Life Sciences"}}],"keywords":[{"id":"https://openalex.org/keywords/5-untranslated-region","display_name":"5' untranslated region","score":0.8700000047683716},{"id":"https://openalex.org/keywords/protein-expression-prediction","display_name":"protein expression prediction","score":0.8140000104904175},{"id":"https://openalex.org/keywords/saccharomyces-cerevisiae","display_name":"Saccharomyces cerevisiae","score":0.9509999752044678},{"id":"https://openalex.org/keywords/kozak-sequence","display_name":"Kozak sequence","score":0.9419999718666077},{"id":"https://openalex.org/keywords/upstream-open-reading-frames","display_name":"upstream open reading frames","score":0.9980000257492065},{"id":"https://openalex.org/keywords/mrna-secondary-structure","display_name":"mRNA secondary structure","score":0.8379999995231628},{"id":"https://openalex.org/keywords/convolutional-neural-networks","display_name":"convolutional neural networks","score":0.828000009059906},{"id":"https://openalex.org/keywords/computational-evolution","display_name":"computational evolution","score":0.6330000162124634},{"id":"https://openalex.org/keywords/synthetic-biology","display_name":"synthetic biology","score":0.6840000152587891},{"id":"https://openalex.org/keywords/gene-design","display_name":"gene design","score":0.6779999732971191}],"concepts":[{"id":"https://openalex.org/C86803240","wikidata":"https://www.wikidata.org/wiki/Q420","display_name":"Biology","level":0,"score":0.8190784454345703},{"id":"https://openalex.org/C89604277","wikidata":"https://www.wikidata.org/wiki/Q424345","display_name":"Untranslated region","level":4,"score":0.6161852478981018},{"id":"https://openalex.org/C2777576037","wikidata":"https://www.wikidata.org/wiki/Q719725","display_name":"Saccharomyces cerevisiae","level":3,"score":0.6111413240432739},{"id":"https://openalex.org/C70721500","wikidata":"https://www.wikidata.org/wiki/Q177005","display_name":"Computational biology","level":1,"score":0.5954817533493042},{"id":"https://openalex.org/C54355233","wikidata":"https://www.wikidata.org/wiki/Q7162","display_name":"Genetics","level":1,"score":0.49185752868652344},{"id":"https://openalex.org/C22667442","wikidata":"https://www.wikidata.org/wiki/Q3272830","display_name":"Three prime untranslated region","level":5,"score":0.46921005845069885},{"id":"https://openalex.org/C21592294","wikidata":"https://www.wikidata.org/wiki/Q3238407","display_name":"Regulatory sequence","level":4,"score":0.4684717059135437},{"id":"https://openalex.org/C81363708","wikidata":"https://www.wikidata.org/wiki/Q17084460","display_name":"Convolutional neural network","level":2,"score":0.46724629402160645},{"id":"https://openalex.org/C12464233","wikidata":"https://www.wikidata.org/wiki/Q937624","display_name":"Five prime untranslated region","level":5,"score":0.4597615599632263},{"id":"https://openalex.org/C165864922","wikidata":"https://www.wikidata.org/wiki/Q411391","display_name":"Regulation of gene expression","level":3,"score":0.37482333183288574},{"id":"https://openalex.org/C104317684","wikidata":"https://www.wikidata.org/wiki/Q7187","display_name":"Gene","level":2,"score":0.36479657888412476},{"id":"https://openalex.org/C154945302","wikidata":"https://www.wikidata.org/wiki/Q11660","display_name":"Artificial intelligence","level":1,"score":0.28764158487319946},{"id":"https://openalex.org/C105580179","wikidata":"https://www.wikidata.org/wiki/Q188928","display_name":"Messenger RNA","level":3,"score":0.20535162091255188},{"id":"https://openalex.org/C41008148","wikidata":"https://www.wikidata.org/wiki/Q21198","display_name":"Computer science","level":0,"score":0.18587461113929749}],"mesh":[{"descriptor_ui":"D000069550","descriptor_name":"Machine Learning","qualifier_ui":null,"qualifier_name":null,"is_major_topic":false},{"descriptor_ui":"D003198","descriptor_name":"Computer Simulation","qualifier_ui":null,"qualifier_name":null,"is_major_topic":false},{"descriptor_ui":"D008957","descriptor_name":"Models, Genetic","qualifier_ui":null,"qualifier_name":null,"is_major_topic":true},{"descriptor_ui":"D012331","descriptor_name":"RNA, Fungal","qualifier_ui":null,"qualifier_name":null,"is_major_topic":false},{"descriptor_ui":"D012333","descriptor_name":"RNA, Messenger","qualifier_ui":null,"qualifier_name":null,"is_major_topic":false},{"descriptor_ui":"D012441","descriptor_name":"Saccharomyces cerevisiae","qualifier_ui":"Q000235","qualifier_name":"genetics","is_major_topic":true},{"descriptor_ui":"D015723","descriptor_name":"Gene Library","qualifier_ui":null,"qualifier_name":null,"is_major_topic":false},{"descriptor_ui":"D016571","descriptor_name":"Neural Networks, Computer","qualifier_ui":null,"qualifier_name":null,"is_major_topic":false},{"descriptor_ui":"D017398","descriptor_name":"Alternative Splicing","qualifier_ui":null,"qualifier_name":null,"is_major_topic":false},{"descriptor_ui":"D020121","descriptor_name":"5' Untranslated Regions","qualifier_ui":null,"qualifier_name":null,"is_major_topic":false}],"locations_count":6,"locations":[{"id":"doi:10.1101/gr.224964.117","doi":"https://doi.org/10.1101/gr.224964.117","is_oa":false,"landing_page_url":"https://doi.org/10.1101/gr.224964.117","pdf_url":null,"source":{"id":"https://openalex.org/S43092948","display_name":"Genome Research","issn_l":"1088-9051","issn":["1088-9051","1549-5469"],"is_oa":false,"is_in_doaj":false,"is_core":true,"listed_in":["anvur-scientific","cwts-core","dongbi-a","doyens","jufo-3","ki-jl-2","medline","norway-2","poland-200","russia-white-list-1"],"host_organization":"https://openalex.org/P4310315909","host_organization_name":"Cold Spring Harbor Laboratory Press","host_organization_lineage":["https://openalex.org/P4310315909"],"host_organization_lineage_names":["Cold Spring Harbor Laboratory Press"],"type":"journal"},"license":null,"license_id":null,"version":"publishedVersion","is_accepted":true,"is_published":true,"raw_source_name":"Genome Research","raw_type":"journal-article"},{"id":"doi:10.1101/137547","doi":"https://doi.org/10.1101/137547","is_oa":true,"landing_page_url":"https://doi.org/10.1101/137547","pdf_url":"https://www.biorxiv.org/content/biorxiv/early/2017/05/19/137547.full.pdf","source":{"id":"https://openalex.org/S4306402567","display_name":"bioRxiv (Cold Spring Harbor Laboratory)","issn_l":null,"issn":null,"is_oa":false,"is_in_doaj":false,"is_core":false,"listed_in":[],"host_organization":"https://openalex.org/I2750212522","host_organization_name":"Cold Spring Harbor Laboratory","host_organization_lineage":["https://openalex.org/I2750212522"],"host_organization_lineage_names":[],"type":"repository"},"license":"cc-by-nc-nd","license_id":"https://openalex.org/licenses/cc-by-nc-nd","version":"acceptedVersion","is_accepted":true,"is_published":false,"raw_source_name":"bioRxiv","raw_type":"posted-content"},{"id":"pmh:oai:europepmc.org:4930535","doi":null,"is_oa":true,"landing_page_url":"https://www.ncbi.nlm.nih.gov/pmc/articles/5741052","pdf_url":null,"source":{"id":"https://openalex.org/S4306400806","display_name":"Europe PMC (PubMed Central)","issn_l":null,"issn":null,"is_oa":false,"is_in_doaj":false,"is_core":false,"listed_in":[],"host_organization":"https://openalex.org/I1303153112","host_organization_name":"European Bioinformatics Institute","host_organization_lineage":["https://openalex.org/I1303153112"],"host_organization_lineage_names":[],"type":"repository"},"license":"cc-by-nc","license_id":"https://openalex.org/licenses/cc-by-nc","version":"publishedVersion","is_accepted":true,"is_published":true,"raw_source_name":null,"raw_type":"Text"},{"id":"pmh:oai:pubmedcentral.nih.gov:5741052","doi":null,"is_oa":true,"landing_page_url":"https://www.ncbi.nlm.nih.gov/pmc/articles/5741052","pdf_url":null,"source":{"id":"https://openalex.org/S2764455111","display_name":"PubMed Central","issn_l":null,"issn":null,"is_oa":true,"is_in_doaj":false,"is_core":false,"listed_in":[],"host_organization":"https://openalex.org/I1299303238","host_organization_name":"National Institutes of Health","host_organization_lineage":["https://openalex.org/I1299303238"],"host_organization_lineage_names":[],"type":"repository"},"license":"cc-by-nc","license_id":"https://openalex.org/licenses/cc-by-nc","version":"publishedVersion","is_accepted":true,"is_published":true,"raw_source_name":"Genome Res","raw_type":"Text"},{"id":"pmid:29097404","doi":null,"is_oa":false,"landing_page_url":"https://pubmed.ncbi.nlm.nih.gov/29097404","pdf_url":null,"source":{"id":"https://openalex.org/S4306525036","display_name":"PubMed","issn_l":null,"issn":null,"is_oa":false,"is_in_doaj":false,"is_core":false,"listed_in":[],"host_organization":"https://openalex.org/I1299303238","host_organization_name":"National Institutes of Health","host_organization_lineage":["https://openalex.org/I1299303238"],"host_organization_lineage_names":[],"type":"repository"},"license":null,"license_id":null,"version":"publishedVersion","is_accepted":true,"is_published":true,"raw_source_name":"Genome research","raw_type":"Journal Article"},{"id":"pmh:oai:open-archive.highwire.org:genome:27/12/2015","doi":null,"is_oa":false,"landing_page_url":"http://genome.cshlp.org/cgi/content/short/27/12/2015","pdf_url":null,"source":{"id":"https://openalex.org/S4406923041","display_name":"HighWire Press Open Archive","issn_l":null,"issn":null,"is_oa":false,"is_in_doaj":false,"is_core":false,"listed_in":[],"host_organization":null,"host_organization_name":null,"host_organization_lineage":[],"host_organization_lineage_names":[],"type":"repository"},"license":null,"license_id":null,"version":"submittedVersion","is_accepted":false,"is_published":false,"raw_source_name":"","raw_type":"TEXT"}],"best_oa_location":{"id":"doi:10.1101/137547","doi":"https://doi.org/10.1101/137547","is_oa":true,"landing_page_url":"https://doi.org/10.1101/137547","pdf_url":"https://www.biorxiv.org/content/biorxiv/early/2017/05/19/137547.full.pdf","source":{"id":"https://openalex.org/S4306402567","display_name":"bioRxiv (Cold Spring Harbor Laboratory)","issn_l":null,"issn":null,"is_oa":false,"is_in_doaj":false,"is_core":false,"listed_in":[],"host_organization":"https://openalex.org/I2750212522","host_organization_name":"Cold Spring Harbor Laboratory","host_organization_lineage":["https://openalex.org/I2750212522"],"host_organization_lineage_names":[],"type":"repository"},"license":"cc-by-nc-nd","license_id":"https://openalex.org/licenses/cc-by-nc-nd","version":"acceptedVersion","is_accepted":true,"is_published":false,"raw_source_name":"bioRxiv","raw_type":"posted-content"},"sustainable_development_goals":[],"study_designs":[],"awards":[{"id":"https://openalex.org/G1861736191","display_name":null,"funder_award_id":"W911NF-11-2-0068","funder_id":"https://openalex.org/F4320332180","funder_display_name":"Defense Advanced Research Projects Agency"},{"id":"https://openalex.org/G4034719064","display_name":"Comprehensive Biology: Exploiting the Yeast Genome","funder_award_id":"P41GM103533","funder_id":"https://openalex.org/F4320337354","funder_display_name":"National Institute of General Medical Sciences"},{"id":"https://openalex.org/G8884853954","display_name":"Collaborative Research: Molecular Programming Architectures, Abstractions, Algorithms, and Applications","funder_award_id":"1317653","funder_id":"https://openalex.org/F4320306076","funder_display_name":"National Science Foundation"},{"id":"https://openalex.org/G963911763","display_name":null,"funder_award_id":"P41 GM103533","funder_id":"https://openalex.org/F4320306076","funder_display_name":"National Science Foundation"}],"funders":[{"id":"https://openalex.org/F4320306076","display_name":"National Science Foundation","ror":"https://ror.org/021nxhr62"},{"id":"https://openalex.org/F4320306082","display_name":"Howard Hughes Medical Institute","ror":"https://ror.org/006w34k90"},{"id":"https://openalex.org/F4320332161","display_name":"National Institutes of Health","ror":"https://ror.org/01cwqze88"},{"id":"https://openalex.org/F4320332180","display_name":"Defense Advanced Research Projects Agency","ror":"https://ror.org/02caytj08"},{"id":"https://openalex.org/F4320332815","display_name":"Advanced Research Projects Agency","ror":"https://ror.org/02caytj08"},{"id":"https://openalex.org/F4320337354","display_name":"National Institute of General Medical Sciences","ror":"https://ror.org/04q48ey07"}],"has_content":{"pdf":true,"grobid_xml":true},"content_urls":{"pdf":"https://content.openalex.org/works/W2951203227.pdf","grobid_xml":"https://content.openalex.org/works/W2951203227.grobid-xml"},"referenced_works_count":80,"referenced_works":["https://openalex.org/W570178133","https://openalex.org/W1019830208","https://openalex.org/W1522301498","https://openalex.org/W1587026990","https://openalex.org/W1601258312","https://openalex.org/W1841597658","https://openalex.org/W1885175685","https://openalex.org/W1966320776","https://openalex.org/W1969995482","https://openalex.org/W1971439989","https://openalex.org/W1976039017","https://openalex.org/W1976144735","https://openalex.org/W1977566832","https://openalex.org/W1979249520","https://openalex.org/W1984948517","https://openalex.org/W1988580221","https://openalex.org/W1989004527","https://openalex.org/W2000137108","https://openalex.org/W2001142396","https://openalex.org/W2003217244","https://openalex.org/W2003993220","https://openalex.org/W2011301426","https://openalex.org/W2017042783","https://openalex.org/W2018645639","https://openalex.org/W2023386301","https://openalex.org/W2027195993","https://openalex.org/W2027368464","https://openalex.org/W2028599082","https://openalex.org/W2038591876","https://openalex.org/W2038847307","https://openalex.org/W2039893699","https://openalex.org/W2044723896","https://openalex.org/W2047360542","https://openalex.org/W2050366196","https://openalex.org/W2053087802","https://openalex.org/W2057677468","https://openalex.org/W2068190648","https://openalex.org/W2070312296","https://openalex.org/W2074007751","https://openalex.org/W2079760969","https://openalex.org/W2080052447","https://openalex.org/W2086561953","https://openalex.org/W2091567823","https://openalex.org/W2097666362","https://openalex.org/W2103534177","https://openalex.org/W2105694586","https://openalex.org/W2107409328","https://openalex.org/W2110621117","https://openalex.org/W2118309386","https://openalex.org/W2119304945","https://openalex.org/W2127667224","https://openalex.org/W2127986230","https://openalex.org/W2134541788","https://openalex.org/W2134887033","https://openalex.org/W2138270253","https://openalex.org/W2139365100","https://openalex.org/W2140240158","https://openalex.org/W2145674897","https://openalex.org/W2146503498","https://openalex.org/W2152052780","https://openalex.org/W2153521373","https://openalex.org/W2154031143","https://openalex.org/W2154241726","https://openalex.org/W2159324718","https://openalex.org/W2159933659","https://openalex.org/W2169456326","https://openalex.org/W2170513170","https://openalex.org/W2198606573","https://openalex.org/W2336509392","https://openalex.org/W2343039591","https://openalex.org/W2345512687","https://openalex.org/W2346609449","https://openalex.org/W2346947737","https://openalex.org/W2403818263","https://openalex.org/W2409674524","https://openalex.org/W2744336424","https://openalex.org/W2765633603","https://openalex.org/W2950980382","https://openalex.org/W2964121744","https://openalex.org/W4230096730"],"related_works":["https://openalex.org/W4404184528","https://openalex.org/W4411168867","https://openalex.org/W3023962428","https://openalex.org/W4379522475","https://openalex.org/W2000137108","https://openalex.org/W4324145726","https://openalex.org/W4393981110","https://openalex.org/W2784340223","https://openalex.org/W3132831899","https://openalex.org/W4399597255"],"abstract_inverted_index":{"Our":[0],"ability":[1],"to":[2,82,145,164],"predict":[3],"protein":[4],"expression":[5],"from":[6],"DNA":[7],"sequence":[8,90],"alone":[9],"remains":[10],"poor,":[11],"reflecting":[12],"our":[13],"limited":[14],"understanding":[15],"of":[16,24,41,47,58,88,122,127,159,176],"cis":[17],"-regulatory":[18],"grammar":[19],"and":[20,66,97,111],"hampering":[21],"the":[22,38,42,50,84,108,119,128,156,160,169,173],"design":[23],"engineered":[25],"genes":[26],"for":[27],"synthetic":[28],"biology":[29],"applications.":[30],"Here,":[31],"we":[32],"generate":[33],"a":[34,56,60,71,102,124],"model":[35,141],"that":[36,113,155],"predicts":[37],"translational":[39,120],"efficiency":[40,121],"5\u2019":[43,64,130,138,150],"untranslated":[44],"region":[45],"(UTR)":[46],"mRNAs":[48],"in":[49,70],"yeast":[51],"Saccharomyces":[52],"cerevisiae.":[53],"We":[54,100,152],"constructed":[55],"library":[57,110],"half":[59],"million":[61],"50-nucleotide-long":[62],"random":[63,109,129],"UTRs":[65,131],"assayed":[67],"their":[68],"activity":[69],"massively":[72],"parallel":[73],"growth":[74],"selection":[75],"experiment.":[76],"The":[77,140],"resulting":[78],"data":[79],"allow":[80],"us":[81],"quantify":[83],"impact":[85],"on":[86,107],"translation":[87,166],"Kozak":[89],"composition,":[91],"upstream":[92],"open":[93],"reading":[94],"frames":[95],"(uORFs)":[96],"secondary":[98],"structure.":[99],"trained":[101],"convolutional":[103],"neural":[104],"network":[105],"(CNN)":[106],"showed":[112],"it":[114],"performs":[115],"well":[116,133],"at":[117],"predicting":[118],"both":[123],"held-out":[125],"set":[126],"as":[132,134],"native":[135],"S.":[136],"cerevisiae":[137],"UTRs.":[139,151],"additionally":[142],"was":[143],"used":[144],"computationally":[146],"evolve":[147],"highly":[148],"translating":[149],"confirmed":[153],"experimentally":[154],"great":[157],"majority":[158],"evolved":[161],"sequences":[162],"lead":[163],"higher":[165],"rates":[167],"than":[168],"starting":[170],"sequences,":[171],"demonstrating":[172],"predictive":[174],"power":[175],"this":[177],"model.":[178]},"counts_by_year":[{"year":2026,"cited_by_count":19},{"year":2025,"cited_by_count":21},{"year":2024,"cited_by_count":25},{"year":2023,"cited_by_count":25},{"year":2022,"cited_by_count":23},{"year":2021,"cited_by_count":27},{"year":2020,"cited_by_count":26},{"year":2019,"cited_by_count":24},{"year":2018,"cited_by_count":14},{"year":2017,"cited_by_count":1}],"updated_date":"2026-11-06T12:05:46.655746","created_date":"2019-06-27T00:00:00"}