From Prompt to Pipeline: Large Language Models for Scientific Workflow Development in Bioinformatics

From Prompt to Pipeline: Large Language Models for Scientific Workflow Development in Bioinformatics | Research Square window.SnipcartSettings = { analytics: { enabled: false } }; (function() { var accessVector = localStorage.getItem('access_vector') || ''; window.dataLayer = window.dataLayer || []; if (accessVector) { window.dataLayer.push({ user: { profile: { profileInfo: { snid: accessVector } } } }); } })(); (function(w,d,s,l,i){w[l]=w[l]||[];w[l].push({'gtm.start':new Date().getTime(),event:'gtm.js'});var f=d.getElementsByTagName(s)[0],j=d.createElement(s),dl=l!='dataLayer'?'&l='+l:'';j.async=true;j.src='https://www.googletagmanager.com/gtm.js?id='+i+dl;f.parentNode.insertBefore(j,f);})(window,document,'script','dataLayer','GTM-K279D39R'); Browse Preprints In Review Journals COVID-19 Preprints AJE Video Bytes Research Tools Research Promotion AJE Professional Editing AJE Rubriq About Preprint Platform In Review Editorial Policies Our Team Advisory Board Help Center Sign In Submit a Preprint Cite Share Download PDF Article From Prompt to Pipeline: Large Language Models for Scientific Workflow Development in Bioinformatics Khairul Alam, Banani Roy This is a preprint; it has not been peer reviewed by a journal. https://doi.org/ 10.21203/rs.3.rs-7642675/v1 This work is licensed under a CC BY 4.0 License Status: Posted Version 1 posted You are reading this latest preprint version Abstract Scientific Workflow Systems (SWSs) such as Galaxy and Nextflow are essential for scalable, reproducible, and automated bioinformatics analyses. However, developing and understanding scientific workflows remains challenging for many domain scientists due to the complexity of tool/module selection, infrastructure requirements, and limited programming expertise. This study explores whether state-of-the-art Large Language Models (LLMs) such as GPT-4o, Gemini 2.5 Flash, and DeepSeek-V3 can assist in generating accurate, complete, and usable bioinformatics workflows. We evaluate a set of representative workflows covering tasks such as RNA-seq, SNP analysis, and DNA methylation across both Galaxy (graphical) and Nextflow (script-based) platforms. To simulate realistic usage, we adopt a tiered prompting strategy: each workflow is first generated using an instruction-only prompt; if the output is incomplete or incorrect, we escalate to a role-based prompt, and finally to chain-of-thought prompting if needed. The generated workflows are evaluated against community-curated baselines from the Galaxy Training Network (GTN) and nf-core, using criteria including correctness, completeness, tool appropriateness, and executability. Results show that LLMs exhibit strong potential in workflow development. Gemini 2.5 Flash produced the most accurate and user-friendly workflows in Galaxy, while DeepSeek-V3 excelled in Nextflow pipeline generation. GPT-4o performed nicely with structured prompts. Prompting strategy significantly influenced output quality, with rolebased and chain-of-thought prompts enhancing correctness and completeness. Overall, LLMs can reduce the cognitive and technical barriers to workflow development, making SWSs more accessible to novice and expert users. This work highlights the practical utility of LLMs and provides actionable insights for integrating them into real-world bioinformatics workflow design. Biological sciences/Computational biology and bioinformatics Physical sciences/Mathematics and computing Scientific Workflows Scientific Workflow Systems Large Language Models Bioinformatics Prompting Techniques Natural Language to Workflow Full Text Additional Declarations No competing interests reported. Cite Share Download PDF Status: Posted Version 1 posted You are reading this latest preprint version Research Square lets you share your work early, gain feedback from the community, and start making changes to your manuscript prior to peer review in a journal. As a division of Research Square Company, we’re committed to making research communication faster, fairer, and more useful. We do this by developing innovative software and high quality services for the global research community. Our growing team is made up of researchers and industry professionals working together to solve the most critical problems facing scientific publishing. Also discoverable on Platform About Our Team In Review Editorial Policies Advisory Board Help Center Resources Author Services Accessibility API Access RSS feed Manage Cookie Preferences © Research Square 2026 | ISSN 2693-5015 (online) Privacy Policy Terms of Service Do Not Sell My Personal Information {"props":{"pageProps":{"initialData":{"identity":"rs-7642675","acceptedTermsAndConditions":true,"allowDirectSubmit":true,"archivedVersions":[],"articleType":"Article","associatedPublications":[],"authors":[{"id":527149351,"identity":"3ede89f6-9a00-41e0-9455-b270b52686b3","order_by":0,"name":"Khairul Alam","email":"data:image/png;base64,iVBORw0KGgoAAAANSUhEUgAAAZAAAAAyAQMAAABI0h/eAAAABlBMVEX///8AAABVwtN+AAAACXBIWXMAAA7EAAAOxAGVKw4bAAAAwElEQVRIiWNgGAWjYBACCQkGhgMfDGxAbGaitTA+nFGRRpoWZmOOM4dJ0CI5u/mZNGPb+cT57ccfGzDU2BHWIi1zzEy6sO124oYzOcYJDMeSCWuRk0gwk54J0sKQw3yAsYEIx8lJpH+T5m07lzi///ljoJZ6IhwmkWNszHPmQGLDjQTjBMaGw4S1SM7IKQQGcrLxhhtvjA0Sjh0nrEXiRvoGYFTayc7vT38s8aGmmrAWVJBAqoZRMApGwSgYBdgBAByQO/XBueqzAAAAAElFTkSuQmCC","orcid":"","institution":"University of Saskatchewan","correspondingAuthor":true,"prefix":"","firstName":"Khairul","middleName":"","lastName":"Alam","suffix":""},{"id":527149352,"identity":"46f80723-3203-4888-99c4-4d2ed2b6133a","order_by":1,"name":"Banani Roy","email":"","orcid":"","institution":"University of Saskatchewan","correspondingAuthor":false,"prefix":"","firstName":"Banani","middleName":"","lastName":"Roy","suffix":""}],"badges":[],"createdAt":"2025-09-17 17:38:12","currentVersionCode":1,"declarations":"","doi":"10.21203/rs.3.rs-7642675/v1","doiUrl":"https://doi.org/10.21203/rs.3.rs-7642675/v1","draftVersion":[],"editorialEvents":[],"editorialNote":"","failedWorkflow":false,"files":[{"id":93271201,"identity":"26d66689-285a-4775-b797-2b605f3b0973","added_by":"auto","created_at":"2025-10-10 23:37:34","extension":"json","order_by":0,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":6469,"visible":true,"origin":"","legend":"","description":"","filename":"d2d4a7bcf8364cc9b2b7c24a28908e86.json","url":"https://assets-eu.researchsquare.com/files/rs-7642675/v1/9555d4c4f677b62b14e006da.json"},{"id":93271159,"identity":"eea5b9c7-6c4d-4537-b05d-5fbceba4d8b6","added_by":"auto","created_at":"2025-10-10 23:29:33","extension":"pdf","order_by":1,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":743221,"visible":true,"origin":"","legend":"","description":"","filename":"PrompttopipelineScientificReports.pdf","url":"https://assets-eu.researchsquare.com/files/rs-7642675/v1/fc44ee4584b6358f541271c1.pdf"},{"id":93271203,"identity":"1bb71607-cefb-47db-824e-01130ef9f8ed","added_by":"auto","created_at":"2025-10-10 23:37:34","extension":"pdf","order_by":2,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":81720,"visible":true,"origin":"","legend":"","description":"","filename":"StudyDesign.pdf","url":"https://assets-eu.researchsquare.com/files/rs-7642675/v1/3b8d8126c1f6aed04a641084.pdf"},{"id":93271160,"identity":"0fe73a7c-a67b-4260-86c2-06a1c9bd5d43","added_by":"auto","created_at":"2025-10-10 23:29:34","extension":"bbl","order_by":3,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":36504,"visible":true,"origin":"","legend":"","description":"","filename":"elsarticletemplatenum.bbl","url":"https://assets-eu.researchsquare.com/files/rs-7642675/v1/6af1c5075162707dee604d72.bbl"},{"id":93271164,"identity":"f7ef6192-559e-4ab2-9ab2-196922f7c2fa","added_by":"auto","created_at":"2025-10-10 23:29:34","extension":"cls","order_by":4,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":44547,"visible":true,"origin":"","legend":"","description":"","filename":"elsarticle.cls","url":"https://assets-eu.researchsquare.com/files/rs-7642675/v1/0c5be80951bcdb7d2b3767f2.cls"},{"id":93271369,"identity":"8894793d-4e79-4982-99e7-76fad16d2e3f","added_by":"auto","created_at":"2025-10-10 23:45:34","extension":"bbl","order_by":5,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":21027,"visible":true,"origin":"","legend":"","description":"","filename":"main.bbl","url":"https://assets-eu.researchsquare.com/files/rs-7642675/v1/aff97fc3a79dd6b2ee274900.bbl"},{"id":93271158,"identity":"ee9ca7de-7374-4cd2-9ad3-28d5f991ab8a","added_by":"auto","created_at":"2025-10-10 23:29:33","extension":"png","order_by":6,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":306439,"visible":true,"origin":"","legend":"","description":"","filename":"nfcorernaseqworkflow.png","url":"https://assets-eu.researchsquare.com/files/rs-7642675/v1/dcfe4fdbbb22310fac4520ec.png"},{"id":93271165,"identity":"a48a51ce-a5fd-4702-8c4c-4b57b77c4bfb","added_by":"auto","created_at":"2025-10-10 23:29:34","extension":"xml","order_by":7,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":265530,"visible":true,"origin":"","legend":"","description":"","filename":"d2d4a7bcf8364cc9b2b7c24a28908e861structuring.xml","url":"https://assets-eu.researchsquare.com/files/rs-7642675/v1/7a62429a3b4e17906669751f.xml"},{"id":95221835,"identity":"455be7c7-7810-4184-b688-1f54b46fa4d1","added_by":"auto","created_at":"2025-11-05 16:19:46","extension":"pdf","order_by":1,"title":"","display":"","copyAsset":false,"role":"manuscript-pdf","size":976912,"visible":true,"origin":"","legend":"","description":"","filename":"PrompttopipelineScientificReports.pdf","url":"https://assets-eu.researchsquare.com/files/rs-7642675/v1_covered_e8690949-3cfb-4b0e-b5b3-6fc9efbb4ece.pdf"}],"financialInterests":"No competing interests reported.","formattedTitle":"From Prompt to Pipeline: Large Language Models for Scientific Workflow Development in Bioinformatics","fulltext":[],"fulltextSource":"","fullText":"","funders":[],"hasAdminPriorityOnWorkflow":false,"hasManuscriptDocX":false,"hasOptedInToPreprint":true,"hasPassedJournalQc":"","hasAnyPriority":false,"hideJournal":true,"highlight":"","institution":"","isAcceptedByJournal":false,"isAuthorSuppliedPdf":true,"isDeskRejected":"","isHiddenFromSearch":false,"isInQc":false,"isInWorkflow":false,"isPdf":true,"isPdfUpToDate":true,"isWithdrawnOrRetracted":false,"journal":{"display":true,"email":"[email protected]","identity":"researchsquare","isNatureJournal":false,"hasQc":true,"allowDirectSubmit":true,"externalIdentity":"","sideBox":"","snPcode":"","submissionUrl":"/submission","title":"Research Square","twitterHandle":"researchsquare","acdcEnabled":true,"dfaEnabled":false,"editorialSystem":"","reportingPortfolio":"","inReviewEnabled":false,"inReviewRevisionsEnabled":true},"keywords":"Scientific Workflows, Scientific Workflow Systems, Large Language Models, Bioinformatics, Prompting Techniques, Natural Language to Workflow","lastPublishedDoi":"10.21203/rs.3.rs-7642675/v1","lastPublishedDoiUrl":"https://doi.org/10.21203/rs.3.rs-7642675/v1","license":{"name":"CC BY 4.0","url":"https://creativecommons.org/licenses/by/4.0/"},"manuscriptAbstract":"\u003cp\u003eScientific Workflow Systems (SWSs) such as Galaxy and Nextflow are essential for scalable, reproducible, and automated bioinformatics analyses. However, developing and understanding scientific workflows remains challenging for many domain scientists due to the complexity of tool/module selection, infrastructure requirements, and limited programming expertise. This study explores whether state-of-the-art Large Language Models (LLMs) such as GPT-4o, Gemini 2.5 Flash, and DeepSeek-V3 can assist in generating accurate, complete, and usable bioinformatics workflows. We evaluate a set of representative workflows covering tasks such as RNA-seq, SNP analysis, and DNA methylation across both Galaxy (graphical) and Nextflow (script-based) platforms. To simulate realistic usage, we adopt a tiered prompting strategy: each workflow is first generated using an instruction-only prompt; if the output is incomplete or incorrect, we escalate to a role-based prompt, and finally to chain-of-thought prompting if needed. The generated workflows are evaluated against community-curated baselines from the Galaxy Training Network (GTN) and nf-core, using criteria including correctness, completeness, tool appropriateness, and executability. Results show that LLMs exhibit strong potential in workflow development. Gemini 2.5 Flash produced the most accurate and user-friendly workflows in Galaxy, while DeepSeek-V3 excelled in Nextflow pipeline generation. GPT-4o performed nicely with structured prompts. Prompting strategy significantly influenced output quality, with rolebased and chain-of-thought prompts enhancing correctness and completeness. Overall, LLMs can reduce the cognitive and technical barriers to workflow development, making SWSs more accessible to novice and expert users. This work highlights the practical utility of LLMs and provides actionable insights for integrating them into real-world bioinformatics workflow design.\u003c/p\u003e","manuscriptTitle":"From Prompt to Pipeline: Large Language Models for Scientific Workflow Development in Bioinformatics","msid":"","msnumber":"","nonDraftVersions":[{"code":1,"date":"2025-10-10 23:29:29","doi":"10.21203/rs.3.rs-7642675/v1","editorialEvents":[{"type":"communityComments","content":0}],"status":"published","journal":{"display":true,"email":"[email protected]","identity":"researchsquare","isNatureJournal":false,"hasQc":true,"allowDirectSubmit":true,"externalIdentity":"","sideBox":"","snPcode":"","submissionUrl":"/submission","title":"Research Square","twitterHandle":"researchsquare","acdcEnabled":true,"dfaEnabled":false,"editorialSystem":"","reportingPortfolio":"","inReviewEnabled":false,"inReviewRevisionsEnabled":true}}],"origin":"","ownerIdentity":"b96abe64-f40a-4a9c-84f1-63feeb7c981a","owner":[],"postedDate":"October 10th, 2025","published":true,"recentEditorialEvents":[],"rejectedJournal":[],"revision":"","amendment":"","status":"posted","subjectAreas":[{"id":56033680,"name":"Biological sciences/Computational biology and bioinformatics"},{"id":56033681,"name":"Physical sciences/Mathematics and computing"}],"tags":[],"updatedAt":"2025-11-03T15:23:48+00:00","versionOfRecord":[],"versionCreatedAt":"2025-10-10 23:29:29","video":"","vorDoi":"","vorDoiUrl":"","workflowStages":[]},"version":"v1","identity":"rs-7642675","journalConfig":"researchsquare"},"__N_SSP":true},"page":"/article/[identity]/[[...version]]","query":{"redirect":"/article/rs-7642675","identity":"rs-7642675","version":["v1"]},"buildId":"8U1c8b4HqxoKbykW_rLl7","isFallback":false,"isExperimentalCompile":false,"dynamicIds":[84888],"gssp":true,"scriptLoader":[]}