{"paper_id":"4e7e8be7-608f-48c7-960e-1ae216d55bba","body_text":"AlphaFold Database Structure Extractor: A web server and API to download AlphaFold structures using common protein accessions | 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 software AlphaFold Database Structure Extractor: A web server and API to download AlphaFold structures using common protein accessions Niharika Saraf, Vishvesh Karthick, Gaurav Sharma This is a preprint; it has not been peer reviewed by a journal. https://doi.org/ 10.21203/rs.3.rs-6583051/v1 This work is licensed under a CC BY 4.0 License Status: Published Journal Publication published 25 Nov, 2025 Read the published version in BMC Bioinformatics → Version 1 posted 4 You are reading this latest preprint version Abstract The AlphaFoldDB Structure Extractor is an open-access web server and API toolkit designed to facilitate the bulk download of predicted protein structures from the AlphaFold Database using well-known accession numbers. Addressing the current limitations in extracting structures beyond a restricted list of model organisms and a threshold number, this tool accepts diverse input identifiers, such as NCBI Taxonomy ID, RefSeq accessions, locus tags, and UniProt or AlphaFold accessions, and maps them to UniProt and AlphaFold IDs for structure retrieval. Users can download structure files in PDB, mmCIF, bCIF, or PAE JSON formats. The tool also generates an accompanying ID mapping file to trace input identifiers back to standard accession numbers and reports unmapped IDs separately. An API methodology is also provided for programmatic access, enabling integration into bioinformatics pipelines. AlphaFoldDB Structure Extractor streamlines the structure procurement process from AlphaFold database, empowering researchers in structural and functional genomics with minimal computational expertise. Availability and implementation: AlphaFoldDB Structure Extractor is freely available as a webserver at https://project.iith.ac.in/sharmaglab/alphafoldextractor/ for downloading structure coordinate files based on the user-provided input of protein accessions. The tool can be accessed programmatically using the API service we have provided. protein structure NCBI UniProt open access AlphaFoldDB Structural genomics Computational biology Bioinformatics API toolkit Figures Figure 1 Figure 2 Introduction The interconnection of protein sequence, structure, and function forms a cornerstone of molecular biology. This sequence-structure-function paradigm emphasizes that the precise arrangement of amino acids dictates the folding patterns and molecular interactions essential for cellular processes. However, this sequence-structure-function relationship is often highly complicated, shaped by evolutionary processes, structural flexibility, and gene neo-functionalization across diverse organisms (Laine et al., 2021 ). Despite this complexity, understanding how sequence variations influence structure and function remains crucial for uncovering biological mechanisms, interpreting biodiversity, and advancing fields such as structural genomics and functional genomics. In the last four decades, the ever-increasing sequence data has laid down the foundation for comparative genomics and propelled the field of structural genomics (Lander, 1996 ) (Standley et al., 2022 ). The first elucidation of a protein structure, i.e., myoglobin by J.C. Kendrew and colleagues in 1958 marked a pivotal advancement (Kendrew et al., 1958 ). As the field developed, the tertiary structures of proteins have been resolved by nuclear magnetic resonance (NMR), X-ray crystallography and cryo-electron microscopy (cryo-EM) which are powerful but labor-intensive techniques and that is why a limited number of experimentally resolved structures had been resolved compared to the availability of sequences (Bertoline et al., 2023 ). Recently, with the use of machine learning and artificial intelligence, there is an avalanche of predicted structures available on the AlphaFold Structure Database (Jumper et al., 2021 ), which has brought a huge paradigm shift in the analysis and interpretation of protein functions. The access to the structure files in PDB, bCIF or mmCIF formats makes it easy for structural biologists and bioinformaticians to perform analyses. AlphaFold DB provides an option for 48 organisms’ bulk structure downloads (model organisms and global health proteome) and otherwise recently, the bulk download has been enabled for up to 100 structures from the webserver. However, there is no direct way on the webserver to download all structure files of organisms other than those 48 organisms. Also, the webserver interface does not allow selectively downloading structures of > 100 proteins via their UniProt or AlphaFold accessions and one can only download the coordinate files programmatically via the AlphaFold API (The UniProt Consortium, 2025). Apart from this, several researchers work with sequence data from NCBI (Sayers et al., 2024 ) and utilize various protein sequence identifiers for their analyses such as RefSeq accessions (O’Leary et al., 2016 ), locus tags, or old locus tags but cannot extract structure files directly. To bridge these gaps, we built AlphaFoldDB Structure Extractor which is a simple web-based tool that lets users download all available predicted structures from AlphaFoldDB for their dataset of interest with just a list of taxonomy/protein accessions as input. Material and methods The AlphaFoldDB Structure Extractor’s workflow has been depicted in Figure 1. Input: AlphaFoldDB Structure Extractor takes the input in several formats including (Figure 2A): NCBI TaxIds – for downloading all available AFDB structure files of proteins of an organism AlphaFold IDs, Uniprot IDs, NCBI RefSeq protein IDs, NCBI locus tags and old locus tags (the latter two are more common for prokaryotic data) – for downloading the structures of a specific list of proteins of interest The structural coordinates can be downloaded in either .pdb, .cif, or .bcif formats and we provide an option to download the predicted aligned error .json files from AlphaFold DB as well. If the user wants to download all file formats, we provide another option for the same. Accession mapping: To retrieve structures from the AlphaFold Structure Database via its FTP page, UniProt accessions are required. If the input is in a different identifier format, conversion to UniProt accessions is necessary. The primary protocol employed by this tool (Figure 1) involves mapping alternative identifiers to UniProt accessions when the query type is not initially in UniProt or AlphaFold format. Uniprot or AlphaFold Accessions: No mapping is required for these inputs and structure files can be downloaded directly from the AlphaFold FTP server via the AlphaFold API. RefSeq Protein Accessions: The RefSeq accessions are mapped to UniProt accessions via the “ID mapping” tool of UniProt and those UniProt accessions are then used as query to download the available AlphaFold predictions. Old Locus Tags or Locus Tags : The old locus tags were used prior to the Prokaryotic Genome Annotation Pipeline (PGAP) re-annotation and the locus tags are the newer annotation. Only the old locus tags are directly recognized on the UniProt database, while the locus tags are not. Hence, if the user provides locus tags, our tool maps them to their corresponding old locus tags using NCBI Entrez Direct by searching in the ‘nuccore’ and ‘gene’ databases. These old locus tags are then mapped to obtain the corresponding UniProt accessions from the UniProt database which are then used to download the available AlphaFold predictions. NCBI TaxIds : When the taxonomy IDs are given as input, our tool sends this string as a query to UniProt from where all the Uniprot accessions corresponding to the proteins belonging to that taxon are obtained and these accessions are then used as a query to download all the available AlphaFold predictions. For all query input types, in some cases, it is possible that when accession mapping is performed, the corresponding UniProt accession is not found or AlphaFold prediction is not available. For such inputs, no structure files will be downloaded, and these inputs will be listed in the output file called ‘Missing_IDs.txt’. Some examples of such cases for each input query type are given on our webserver’s ‘Help’ page. Computational Methods: The tool was written in shell script and has a dependency of Entrez Direct command-line utilities (https://www.ncbi.nlm.nih.gov/books/NBK179288/) for the accession mapping of locus tags. “Uniprot ID mapping” was utilized programmatically for RefSeq accession to UniProt accession mapping. For developing the webserver, the shell script was spawned as a child process using node.js. The webserver was developed using HTML, CSS and JavaScript for the frontend and Node.Js - javascript runtime environment, BASH for the backend. We have also developed programmatic access to our tool via the AFDB Structure Extractor API, which has been explained with examples on our webserver’s ‘Help’ page (https://project.iith.ac.in/sharmaglab/alphafoldextractor/help.html). Results The user should select any query type first and then paste all identifiers of their interest in the input section. The server maps all input identifiers to UniProt, AlphaFold and NCBI accessions as per the user query type and retrieves the corresponding structural data. Upon completion, users can download a compressed (zipped) file from the web server. This zipped archive contains the structure files in multiple formats, PDB, mmCIF, bCIF, and/or predicted aligned error JSON files, organized and named according to their AlphaFold accessions (Figure 2B). Along with the structure files, the output package also contains an ‘ID_mapping.csv’ file, which allows the users to trace back their IDs in different formats (Figure 2C). This csv file contains input accessions along with their mapped UniProt and other accessions, in cases where the mapping was performed. The list of accessions for which the corresponding old locus tags (when locus tags are the input), or UniProt accessions or AlphaFold accessions are not found, they get listed in the ‘Missing_IDs.txt’ file. Furthermore, for users submitting a large number of query accessions, preferably >50 accessions, an option to provide an email address is available. Upon request, the system will automatically send a downloadable link to the user via email, granting access to the zipped archive containing the retrieved structures and associated ID mapping information. For each session, user will be able to see all runs and their results in the bottom of the input page, allowing users to download the structures of accession of their interest. API-based services: This service will allow the expert users to download structures on their terminal along with incorporating these commands in their own pipelines. Template URL for single accession as input: https://project.iith.ac.in/sharmaglab/alphafoldextractor/api/<id-type>/<id> Template URL for multiple accessions as input: https://project.iith.ac.in/sharmaglab/alphafoldextractor/api/<id-type> <id-type> : oldlocustag / locustag / uniprot / alphafold / refseq / taxonomy For multiple accessions, only ids and optionally format are accepted in request body. format deafaults to 'pdb'. ids : Comma seperated values. Spaces and '_' are tolerated. Unsupported symbols are converted to '_'. format : [ pdb ] / cif / bcif / pae / all Examples: Single accession as input: curl -JL https://project.iith.ac.in/sharmaglab/alphafoldextractor/api/oldlocustag/MXAN_1028 > structures.zip wget --output-document=structures.zip https://project.iith.ac.in/sharmaglab/alphafoldextractor/api/oldlocustag/MXAN_1028 Multiple accessions as input: curl -OJL -d \"ids=MXAN_1028,GVO57_04600\" https://project.iith.ac.in/sharmaglab/alphafoldextractor/api/oldlocustag curl -JL -d \"ids=MXAN_1028,GVO57_04600&&format=bcif\" https://project.iith.ac.in/sharmaglab/alphafoldextractor/api/oldlocustag > MXAN_1028.zip curl -JL -d \"ids=MXAN_1028,GVO57_04600\" -d \"format=bcif\" https://project.iith.ac.in/sharmaglab/alphafoldextractor/api/oldlocustag > MXAN_1028.zip wget --post-data=\"ids=MXAN_1028,GVO57_04600\" --content-disposition --trust-server-names https://project.iith.ac.in/sharmaglab/alphafoldextractor/api/oldlocustag wget --post-data=\"ids=MXAN_1028&format=bcif\" --output-document=MXAN_1028.zip https://project.iith.ac.in/sharmaglab/alphafoldextractor/api/oldlocustag Limitations Currently, we provide the input formats of only some widely used accessions specific to NCBI, but there are several different accession types used by different platforms. We plan to extend the input formats in future. Conclusion We have developed AlphaFoldDB Structure Extractor for bulk downloading AlphaFold structure predictions easily based on different accessions/identifiers as a contribution towards the advancement of protein structure analysis. Our aim in developing AFDB Structure Extractor was to resolve identifiers from different sources into their corresponding UniProt and AlphaFold accessions while providing an easy-to-use bloat-free interface along with REST API accessible endpoints. This tool will majorly help experimental researchers who are not quite familiar with the computational knowledge to extract AlphaFold structures for their proteins of interest, further allowing them to work on structural genomics aspects. It will also potentially help structural biologists and bioinformaticians to streamline AlphaFold structure predictions into their pipelines without having to worry about intermediate identifier resolution. Declarations Acknowledgements : We thank PARAM Seva, a National Supercomputing facility within IIT Hyderabad for their support in our research. Ethics approval and consent to participate: Not applicable Availability of data and materials: Authors have used open-source tools in this analysis, for which additional information has been provided at https://project.iith.ac.in/sharmaglab/alphafoldextractor/. Competing interests: The authors declare no conflict of interest to disclose. Funding: GS acknowledges the seed grant from IIT Hyderabad and Start-up Research Grant (SRG) from Science and Engineering Research Board (SERB) for supporting his research. NS and VK thank the fellowship support by IIT Hyderabad and Science and Engineering Research Board (SERB) respectively. Authors' contributions: GS conceptualized the idea and supervised the project. NS performed the major analysis and wrote the first draft. VK implemented the back-end development. NS, VK, and GS edited and finalized the manuscript. All authors approved the final version. References Bertoline,L.M.F. et al. (2023) Before and after AlphaFold2: An overview of protein structure prediction. Frontiers in Bioinformatics, 3, 1120370. Consortium,T.U. et al. (2025) UniProt: the Universal Protein Knowledgebase in 2025. Nucleic Acids Res , 53, D609–D617. Jumper,J. et al. (2021) Highly accurate protein structure prediction with AlphaFold. Nature 2021 596:7873 , 596, 583–589. Kendrew,J.C. et al. (1958) A Three-Dimensional Model of the Myoglobin Molecule Obtained by X-Ray Analysis. Nature 1958 181:4610 , 181, 662–666. Laine E, et al . (2021) Protein sequence-to-structure learning: Is this the end (-to-end revolution)?. Proteins. 89(12):1770-1786. Lander,E.S. (1996) The New Genomics: Global Views of Biology. Science (1979) , 274, 536–539. O’Leary,N.A. et al. (2016) Reference sequence (RefSeq) database at NCBI: current status, taxonomic expansion, and functional annotation. Nucleic Acids Res , 44, D733–D745. Sayers,E.W. et al. (2024) Database resources of the National Center for Biotechnology Information. Nucleic Acids Res , 52, D33–D43. Standley,D.M. et al. (2022) The evolution of structural genomics. Biophys Rev , 14, 1247. Additional Declarations No competing interests reported. Cite Share Download PDF Status: Published Journal Publication published 25 Nov, 2025 Read the published version in BMC Bioinformatics → Version 1 posted Editorial decision: Revision requested 07 May, 2025 Editor assigned by journal 05 May, 2025 Submission checks completed at journal 05 May, 2025 First submitted to journal 03 May, 2025 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-6583051\",\"acceptedTermsAndConditions\":true,\"allowDirectSubmit\":false,\"archivedVersions\":[],\"articleType\":\"software\",\"associatedPublications\":[],\"authors\":[{\"id\":451954008,\"identity\":\"9a34c006-7cad-4ee1-b0bd-194d855aa51e\",\"order_by\":0,\"name\":\"Niharika Saraf\",\"email\":\"\",\"orcid\":\"\",\"institution\":\"Indian Institute of Technology Hyderabad\",\"correspondingAuthor\":false,\"prefix\":\"\",\"firstName\":\"Niharika\",\"middleName\":\"\",\"lastName\":\"Saraf\",\"suffix\":\"\"},{\"id\":451954011,\"identity\":\"b790e5cb-6c36-4ff7-a1fd-64d9f430b60a\",\"order_by\":1,\"name\":\"Vishvesh Karthick\",\"email\":\"\",\"orcid\":\"\",\"institution\":\"Indian Institute of Technology Hyderabad\",\"correspondingAuthor\":false,\"prefix\":\"\",\"firstName\":\"Vishvesh\",\"middleName\":\"\",\"lastName\":\"Karthick\",\"suffix\":\"\"},{\"id\":451954013,\"identity\":\"904e7976-4483-4517-b48a-7192861d5ac1\",\"order_by\":2,\"name\":\"Gaurav Sharma\",\"email\":\"data:image/png;base64,iVBORw0KGgoAAAANSUhEUgAAAZAAAAAyAQMAAABI0h/eAAAABlBMVEX///8AAABVwtN+AAAACXBIWXMAAA7EAAAOxAGVKw4bAAAA3klEQVRIie2OMQrCQBBFJyykCtimSq6wIYWNeJYNAdNsQLC1CAixsxa8hGAh6QYGkiYHSGFjY68BsRLNgtptBBuLfcWfGZgHH8Bg+EdclWOPPxMFoDrxCyUOud0p4nsFo63dTaF/VvibxenSzDHZ+5mFx9sBBku0aKpR+KEMN7LEtMhVsRO4tQBa6xRXhEzaVbotlUIADQA5umLrpGXyTgl/KX6fAo0MWZqTeCu8T+GNnLF0FQdFHmUoJuQEdZT1FdsxeR37Q0Z0vo3I8yqiVlvsg5V16bwWg8FgMPzAA2riU/jlGhfLAAAAAElFTkSuQmCC\",\"orcid\":\"\",\"institution\":\"Indian Institute of Technology Hyderabad\",\"correspondingAuthor\":true,\"prefix\":\"\",\"firstName\":\"Gaurav\",\"middleName\":\"\",\"lastName\":\"Sharma\",\"suffix\":\"\"}],\"badges\":[],\"createdAt\":\"2025-05-03 08:53:18\",\"currentVersionCode\":1,\"declarations\":\"\",\"doi\":\"10.21203/rs.3.rs-6583051/v1\",\"doiUrl\":\"https://doi.org/10.21203/rs.3.rs-6583051/v1\",\"draftVersion\":[],\"editorialEvents\":[{\"content\":\"https://doi.org/10.1186/s12859-025-06303-0\",\"type\":\"published\",\"date\":\"2025-11-25T15:57:00+00:00\"}],\"editorialNote\":\"\",\"failedWorkflow\":false,\"files\":[{\"id\":82083813,\"identity\":\"5439fb67-6bd4-4384-8f99-bb4210e1f6ca\",\"added_by\":\"auto\",\"created_at\":\"2025-05-06 14:52:32\",\"extension\":\"png\",\"order_by\":1,\"title\":\"Figure 1\",\"display\":\"\",\"copyAsset\":false,\"role\":\"figure\",\"size\":150030,\"visible\":true,\"origin\":\"\",\"legend\":\"\\u003cp\\u003e\\u003cstrong\\u003eAlphaFoldDB Structure Extractor pipeline for downloading structures using diverse ID types\\u003c/strong\\u003e. The input query options are mentioned on the left followed by the accession mapping pipeline carried out in the backend which uses diverse tools for mapping the different input query accessions to their corresponding UniProt accessions. The final step is downloading the structure coordinate files from the AlphaFold Database FTP server.\\u003c/p\\u003e\",\"description\":\"\",\"filename\":\"1.png\",\"url\":\"https://assets-eu.researchsquare.com/files/rs-6583051/v1/d4d9d843852d63751abdd052.png\"},{\"id\":82083811,\"identity\":\"be8d0024-f710-4073-a78e-710a442e78d6\",\"added_by\":\"auto\",\"created_at\":\"2025-05-06 14:52:32\",\"extension\":\"png\",\"order_by\":2,\"title\":\"Figure 2\",\"display\":\"\",\"copyAsset\":false,\"role\":\"figure\",\"size\":249791,\"visible\":true,\"origin\":\"\",\"legend\":\"\\u003cp\\u003e\\u003cstrong\\u003eWeb interface of AlphaFoldDB Structure Extractor. A) \\u003c/strong\\u003eSnapshot of the web interface. The users must first select their choice of input query type and output file format(s). Then they must submit their query accessions as a text input or a text file with each accession in a new line. The example given in the figure is to download all file formats with locus tags as input. The users can optionally provide their email address which is recommended for more than 50 query inputs. \\u003cstrong\\u003eB) \\u003c/strong\\u003eOnce all the accessions are processed, the output will be available in the form of a link to download the zip file having all the structure coordinate files named as per the AlphaFold accessions. The zip file also has the ID mapping file and missing IDs file. The download link is sent via email, if the user has provided their email-address. \\u003cstrong\\u003eC)\\u003c/strong\\u003e An ID mapping file is also provided from which the user can trace back each AlphaFold accession’s corresponding input query accession for ease of identification and analysis. In the case of locus tags as shown, an additional step of mapping to old locus tags is also carried out.\\u003c/p\\u003e\",\"description\":\"\",\"filename\":\"2.png\",\"url\":\"https://assets-eu.researchsquare.com/files/rs-6583051/v1/bf73dc1d5d43525c50caed4a.png\"},{\"id\":97178237,\"identity\":\"bff4c90e-c767-47e0-b6f9-d5adb0af3faf\",\"added_by\":\"auto\",\"created_at\":\"2025-12-01 16:04:00\",\"extension\":\"pdf\",\"order_by\":0,\"title\":\"\",\"display\":\"\",\"copyAsset\":false,\"role\":\"manuscript-pdf\",\"size\":793935,\"visible\":true,\"origin\":\"\",\"legend\":\"\",\"description\":\"\",\"filename\":\"manuscript.pdf\",\"url\":\"https://assets-eu.researchsquare.com/files/rs-6583051/v1/7ddb8eb7-2733-4a67-92a9-5e8bd8290719.pdf\"}],\"financialInterests\":\"No competing interests reported.\",\"formattedTitle\":\"AlphaFold Database Structure Extractor: A web server and API to download AlphaFold structures using common protein accessions\",\"fulltext\":[{\"header\":\"Introduction\",\"content\":\"\\u003cp\\u003eThe interconnection of protein sequence, structure, and function forms a cornerstone of molecular biology. This sequence-structure-function paradigm emphasizes that the precise arrangement of amino acids dictates the folding patterns and molecular interactions essential for cellular processes. However, this sequence-structure-function relationship is often highly complicated, shaped by evolutionary processes, structural flexibility, and gene neo-functionalization across diverse organisms (Laine et al., \\u003cspan citationid=\\\"CR5\\\" class=\\\"CitationRef\\\"\\u003e2021\\u003c/span\\u003e). Despite this complexity, understanding how sequence variations influence structure and function remains crucial for uncovering biological mechanisms, interpreting biodiversity, and advancing fields such as structural genomics and functional genomics.\\u003c/p\\u003e \\u003cp\\u003eIn the last four decades, the ever-increasing sequence data has laid down the foundation for comparative genomics and propelled the field of structural genomics (Lander, \\u003cspan citationid=\\\"CR6\\\" class=\\\"CitationRef\\\"\\u003e1996\\u003c/span\\u003e) (Standley et al., \\u003cspan citationid=\\\"CR9\\\" class=\\\"CitationRef\\\"\\u003e2022\\u003c/span\\u003e). The first elucidation of a protein structure, i.e., myoglobin by J.C. Kendrew and colleagues in 1958 marked a pivotal advancement (Kendrew et al., \\u003cspan citationid=\\\"CR4\\\" class=\\\"CitationRef\\\"\\u003e1958\\u003c/span\\u003e). As the field developed, the tertiary structures of proteins have been resolved by nuclear magnetic resonance (NMR), X-ray crystallography and cryo-electron microscopy (cryo-EM) which are powerful but labor-intensive techniques and that is why a limited number of experimentally resolved structures had been resolved compared to the availability of sequences (Bertoline et al., \\u003cspan citationid=\\\"CR1\\\" class=\\\"CitationRef\\\"\\u003e2023\\u003c/span\\u003e).\\u003c/p\\u003e \\u003cp\\u003eRecently, with the use of machine learning and artificial intelligence, there is an avalanche of predicted structures available on the AlphaFold Structure Database (Jumper et al., \\u003cspan citationid=\\\"CR3\\\" class=\\\"CitationRef\\\"\\u003e2021\\u003c/span\\u003e), which has brought a huge paradigm shift in the analysis and interpretation of protein functions. The access to the structure files in PDB, bCIF or mmCIF formats makes it easy for structural biologists and bioinformaticians to perform analyses. AlphaFold DB provides an option for 48 organisms\\u0026rsquo; bulk structure downloads (model organisms and global health proteome) and otherwise recently, the bulk download has been enabled for up to 100 structures from the webserver. However, there is no direct way on the webserver to download all structure files of organisms other than those 48 organisms. Also, the webserver interface does not allow selectively downloading structures of \\u0026gt;\\u0026thinsp;100 proteins via their UniProt or AlphaFold accessions and one can only download the coordinate files programmatically via the AlphaFold API (The UniProt Consortium, 2025). Apart from this, several researchers work with sequence data from NCBI (Sayers et al., \\u003cspan citationid=\\\"CR8\\\" class=\\\"CitationRef\\\"\\u003e2024\\u003c/span\\u003e) and utilize various protein sequence identifiers for their analyses such as RefSeq accessions (O\\u0026rsquo;Leary et al., \\u003cspan citationid=\\\"CR7\\\" class=\\\"CitationRef\\\"\\u003e2016\\u003c/span\\u003e), locus tags, or old locus tags but cannot extract structure files directly.\\u003c/p\\u003e \\u003cp\\u003eTo bridge these gaps, we built AlphaFoldDB Structure Extractor which is a simple web-based tool that lets users download all available predicted structures from AlphaFoldDB for their dataset of interest with just a list of taxonomy/protein accessions as input.\\u003c/p\\u003e\"},{\"header\":\"Material and methods\",\"content\":\"\\u003cp\\u003eThe AlphaFoldDB Structure Extractor\\u0026rsquo;s workflow has been depicted in Figure 1.\\u0026nbsp;\\u003c/p\\u003e\\n\\u003cp\\u003e\\u003cstrong\\u003eInput:\\u0026nbsp;\\u003c/strong\\u003eAlphaFoldDB Structure Extractor takes the input in several formats including (Figure 2A):\\u003c/p\\u003e\\n\\u003col\\u003e\\n \\u003cli\\u003eNCBI TaxIds \\u0026ndash; for downloading all available AFDB structure files of proteins of an organism\\u003c/li\\u003e\\n \\u003cli\\u003eAlphaFold IDs, Uniprot IDs, NCBI RefSeq protein IDs, NCBI locus tags and old locus tags (the latter two are more common for prokaryotic data) \\u0026ndash; for downloading the structures of a specific list of proteins of interest\\u003c/li\\u003e\\n\\u003c/ol\\u003e\\n\\u003cp\\u003eThe structural coordinates can be downloaded in either .pdb, .cif, or .bcif formats and we provide an option to download the predicted aligned error .json files from AlphaFold DB as well. If the user wants to download all file formats, we provide another option for the same.\\u0026nbsp;\\u003c/p\\u003e\\n\\u003cp\\u003e\\u003cstrong\\u003eAccession mapping:\\u0026nbsp;\\u003c/strong\\u003eTo retrieve structures from the AlphaFold Structure Database via its FTP page, UniProt accessions are required. If the input is in a different identifier format, conversion to UniProt accessions is necessary. The primary protocol employed by this tool (Figure 1) involves mapping alternative identifiers to UniProt accessions when the query type is not initially in UniProt or AlphaFold format.\\u003c/p\\u003e\\n\\u003cp\\u003e\\u003cem\\u003eUniprot or AlphaFold Accessions:\\u0026nbsp;\\u003c/em\\u003eNo mapping is required for these inputs and structure files can be downloaded directly from the AlphaFold FTP server via the AlphaFold API.\\u0026nbsp;\\u003c/p\\u003e\\n\\u003cp\\u003e\\u003cem\\u003eRefSeq Protein Accessions:\\u0026nbsp;\\u003c/em\\u003eThe RefSeq accessions are mapped to UniProt accessions via the \\u0026ldquo;ID mapping\\u0026rdquo; tool of UniProt and those UniProt accessions are then used as query to download the available AlphaFold predictions.\\u0026nbsp;\\u003c/p\\u003e\\n\\u003cp\\u003e\\u003cem\\u003eOld Locus Tags or Locus Tags\\u003c/em\\u003e: The old locus tags were used prior to the Prokaryotic Genome Annotation Pipeline (PGAP) re-annotation and the locus tags are the newer annotation. Only the old locus tags are directly recognized on the UniProt database, while the locus tags are not. Hence, if the user provides locus tags, our tool maps them to their corresponding old locus tags using NCBI Entrez Direct by searching in the \\u0026lsquo;nuccore\\u0026rsquo; and \\u0026lsquo;gene\\u0026rsquo; databases. These old locus tags are then mapped to obtain the corresponding UniProt accessions from the UniProt database which are then used to download the available AlphaFold predictions.\\u003c/p\\u003e\\n\\u003cp\\u003e\\u003cem\\u003eNCBI TaxIds\\u003c/em\\u003e: When the taxonomy IDs are given as input, our tool sends this string as a query to UniProt from where all the Uniprot accessions corresponding to the proteins belonging to that taxon are obtained and these accessions are then used as a query to download all the available AlphaFold predictions.\\u003c/p\\u003e\\n\\u003cp\\u003eFor all query input types, in some cases, it is possible that when accession mapping is performed, the corresponding UniProt accession is not found or AlphaFold prediction is not available. For such inputs, no structure files will be downloaded, and these inputs will be listed in the output file called \\u0026lsquo;Missing_IDs.txt\\u0026rsquo;. Some examples of such cases for each input query type are given on our webserver\\u0026rsquo;s \\u0026lsquo;Help\\u0026rsquo; page. \\u0026nbsp;\\u003c/p\\u003e\\n\\u003cp\\u003e\\u003cstrong\\u003eComputational Methods:\\u0026nbsp;\\u003c/strong\\u003eThe tool was written in shell script and has a dependency of Entrez Direct command-line utilities (https://www.ncbi.nlm.nih.gov/books/NBK179288/) for the accession mapping of locus tags. \\u0026ldquo;Uniprot ID mapping\\u0026rdquo; was utilized programmatically for RefSeq accession to UniProt accession mapping. For developing the webserver, the shell script was spawned as a child process using node.js. The webserver was developed using HTML, CSS and JavaScript for the frontend and Node.Js - javascript runtime environment, BASH for the backend. We have also developed programmatic access to our tool via the AFDB Structure Extractor API, which has been explained with examples on our webserver\\u0026rsquo;s \\u0026lsquo;Help\\u0026rsquo; page (https://project.iith.ac.in/sharmaglab/alphafoldextractor/help.html).\\u003c/p\\u003e\"},{\"header\":\"Results\",\"content\":\"\\u003cp\\u003eThe user should select any query type first and then paste all identifiers of their interest in the input section. The server maps all input identifiers to UniProt, AlphaFold and NCBI accessions as per the user query type and retrieves the corresponding structural data. Upon completion, users can download a compressed (zipped) file from the web server. This zipped archive contains the structure files in multiple formats, PDB, mmCIF, bCIF, and/or predicted aligned error JSON files, organized and named according to their AlphaFold accessions (Figure 2B).\\u003c/p\\u003e\\n\\u003cp\\u003eAlong with the structure files, the output package also contains an ‘ID_mapping.csv’ file, which allows the users to trace back their IDs in different formats (Figure 2C). This csv file contains input accessions along with their mapped UniProt and other accessions, in cases where the mapping was performed. The list of accessions for which the corresponding old locus tags (when locus tags are the input), or UniProt accessions or AlphaFold accessions are not found, they get listed in the ‘Missing_IDs.txt’ file.\\u0026nbsp;\\u003c/p\\u003e\\n\\u003cp\\u003eFurthermore, for users submitting a large number of query accessions, preferably \\u0026gt;50 accessions, an option to provide an email address is available. Upon request, the system will automatically send a downloadable link to the user via email, granting access to the zipped archive containing the retrieved structures and associated ID mapping information.\\u0026nbsp;\\u003c/p\\u003e\\n\\u003cp\\u003eFor each session, user will be able to see all runs and their results in the bottom of the input page, allowing users to download the structures of accession of their interest.\\u0026nbsp;\\u003c/p\\u003e\\n\\u003cp\\u003e\\u003cstrong\\u003eAPI-based services:\\u0026nbsp;\\u003c/strong\\u003e\\u003c/p\\u003e\\n\\u003cp\\u003eThis service will allow the expert users to download structures on their terminal along with incorporating these commands in their own pipelines.\\u003c/p\\u003e\\n\\u003cp\\u003eTemplate URL for single accession as input:\\u0026nbsp;\\u003c/p\\u003e\\n\\u003cul\\u003e\\n \\u003cli\\u003ehttps://project.iith.ac.in/sharmaglab/alphafoldextractor/api/\\u0026lt;id-type\\u0026gt;/\\u0026lt;id\\u0026gt;\\u003c/li\\u003e\\n\\u003c/ul\\u003e\\n\\u003cp\\u003eTemplate URL for multiple accessions as input:\\u0026nbsp;\\u003c/p\\u003e\\n\\u003cul\\u003e\\n \\u003cli\\u003ehttps://project.iith.ac.in/sharmaglab/alphafoldextractor/api/\\u0026lt;id-type\\u0026gt;\\u003c/li\\u003e\\n\\u003c/ul\\u003e\\n\\u003cp\\u003e\\u003cstrong\\u003e\\u0026lt;id-type\\u0026gt;\\u003c/strong\\u003e: \\u003cem\\u003eoldlocustag / locustag / uniprot / alphafold / refseq / taxonomy\\u003c/em\\u003e\\u003cbr\\u003e\\u0026nbsp;\\u003cbr\\u003eFor multiple accessions, only \\u003cem\\u003eids\\u003c/em\\u003e and optionally \\u003cem\\u003eformat\\u003c/em\\u003e are accepted in request body. \\u003cem\\u003eformat\\u003c/em\\u003e deafaults to 'pdb'.\\u003c/p\\u003e\\n\\u003cp\\u003e\\u003cstrong\\u003eids\\u003c/strong\\u003e: Comma seperated values. Spaces and '_' are tolerated. Unsupported symbols are converted to '_'.\\u003c/p\\u003e\\n\\u003cp\\u003e\\u003cstrong\\u003eformat\\u003c/strong\\u003e: \\u003cem\\u003e[\\u003cstrong\\u003epdb\\u003c/strong\\u003e] / cif / bcif / pae / all\\u003c/em\\u003e\\u003c/p\\u003e\\n\\u003ch3\\u003eExamples:\\u003c/h3\\u003e\\n\\u003ch4\\u003eSingle accession as input:\\u003c/h4\\u003e\\n\\u003cul\\u003e\\n \\u003cli\\u003ecurl -JL https://project.iith.ac.in/sharmaglab/alphafoldextractor/api/oldlocustag/MXAN_1028 \\u0026gt; structures.zip\\u003c/li\\u003e\\n \\u003cli\\u003ewget --output-document=structures.zip https://project.iith.ac.in/sharmaglab/alphafoldextractor/api/oldlocustag/MXAN_1028\\u003c/li\\u003e\\n\\u003c/ul\\u003e\\n\\u003ch4\\u003eMultiple accessions as input:\\u003c/h4\\u003e\\n\\u003cul\\u003e\\n \\u003cli\\u003ecurl -OJL -d \\\"ids=MXAN_1028,GVO57_04600\\\"\\u0026nbsp;https://project.iith.ac.in/sharmaglab/alphafoldextractor/api/oldlocustag\\u003c/li\\u003e\\n \\u003cli\\u003ecurl -JL -d \\\"ids=MXAN_1028,GVO57_04600\\u0026amp;\\u0026amp;format=bcif\\\" https://project.iith.ac.in/sharmaglab/alphafoldextractor/api/oldlocustag \\u0026gt; MXAN_1028.zip\\u003c/li\\u003e\\n \\u003cli\\u003ecurl -JL -d \\\"ids=MXAN_1028,GVO57_04600\\\" -d \\\"format=bcif\\\" https://project.iith.ac.in/sharmaglab/alphafoldextractor/api/oldlocustag \\u0026gt; MXAN_1028.zip\\u003c/li\\u003e\\n \\u003cli\\u003ewget --post-data=\\\"ids=MXAN_1028,GVO57_04600\\\" --content-disposition --trust-server-names\\u0026nbsp;https://project.iith.ac.in/sharmaglab/alphafoldextractor/api/oldlocustag\\u003c/li\\u003e\\n \\u003cli\\u003ewget --post-data=\\\"ids=MXAN_1028\\u0026amp;format=bcif\\\" --output-document=MXAN_1028.zip https://project.iith.ac.in/sharmaglab/alphafoldextractor/api/oldlocustag\\u003c/li\\u003e\\n\\u003c/ul\\u003e\\n\\u003cp\\u003e\\u003cstrong\\u003eLimitations\\u003c/strong\\u003e\\u003c/p\\u003e\\n\\u003cp\\u003eCurrently, we provide the input formats of only some widely used accessions specific to NCBI, but there are several different accession types used by different platforms. We plan to extend the input formats in future.\\u003c/p\\u003e\"},{\"header\":\"Conclusion\",\"content\":\"\\u003cp\\u003eWe have developed AlphaFoldDB Structure Extractor for bulk downloading AlphaFold structure predictions easily based on different accessions/identifiers as a contribution towards the advancement of protein structure analysis. Our aim in developing AFDB Structure Extractor was to resolve identifiers from different sources into their corresponding UniProt and AlphaFold accessions while providing an easy-to-use bloat-free interface along with REST API accessible endpoints. This tool will majorly help experimental researchers who are not quite familiar with the computational knowledge to extract AlphaFold structures for their proteins of interest, further allowing them to work on structural genomics aspects. It will also potentially help structural biologists and bioinformaticians to streamline AlphaFold structure predictions into their pipelines without having to worry about intermediate identifier resolution.\\u003c/p\\u003e\"},{\"header\":\"Declarations\",\"content\":\"\\u003cp\\u003e\\u003cstrong\\u003eAcknowledgements\\u003c/strong\\u003e: We thank PARAM Seva, a National Supercomputing facility within IIT Hyderabad for their support in our research.\\u003c/p\\u003e\\n\\u003cp\\u003e\\u003cstrong\\u003eEthics approval and consent to participate:\\u0026nbsp;\\u003c/strong\\u003eNot applicable\\u003c/p\\u003e\\n\\u003cp\\u003e\\u003cstrong\\u003eAvailability of data and materials:\\u0026nbsp;\\u003c/strong\\u003eAuthors have used open-source tools in this analysis, for which additional information has been provided at https://project.iith.ac.in/sharmaglab/alphafoldextractor/.\\u0026nbsp;\\u003c/p\\u003e\\n\\u003cp\\u003e\\u003cstrong\\u003eCompeting interests:\\u0026nbsp;\\u003c/strong\\u003eThe authors declare no conflict of interest to disclose.\\u003c/p\\u003e\\n\\u003cp\\u003e\\u003cstrong\\u003eFunding:\\u0026nbsp;\\u003c/strong\\u003eGS acknowledges the seed grant from IIT Hyderabad and Start-up Research Grant (SRG) from\\u0026nbsp;Science and Engineering Research Board (SERB) for supporting his research. NS and VK thank the fellowship support by IIT Hyderabad and Science and Engineering Research Board (SERB) respectively.\\u0026nbsp;\\u003c/p\\u003e\\n\\u003cp\\u003e\\u003cstrong\\u003eAuthors\\u0026apos; contributions:\\u0026nbsp;\\u003c/strong\\u003eGS conceptualized the idea and supervised the project. NS performed the major analysis and wrote the first draft. VK implemented the back-end development. NS, VK, and GS edited and finalized the manuscript. All authors approved the final version.\\u0026nbsp;\\u003c/p\\u003e\"},{\"header\":\"References\",\"content\":\"\\u003col\\u003e\\n\\u003cli\\u003e\\u003cem\\u003eBertoline,L.M.F. et al. (2023) Before and after AlphaFold2: An overview of protein structure prediction. Frontiers in Bioinformatics, 3, 1120370.\\u003c/em\\u003e\\u003c/li\\u003e\\n\\u003cli\\u003eConsortium,T.U. \\u003cem\\u003eet al.\\u003c/em\\u003e (2025) UniProt: the Universal Protein Knowledgebase in 2025. \\u003cem\\u003eNucleic Acids Res\\u003c/em\\u003e, 53, D609\\u0026ndash;D617.\\u003c/li\\u003e\\n\\u003cli\\u003eJumper,J. \\u003cem\\u003eet al.\\u003c/em\\u003e (2021) Highly accurate protein structure prediction with AlphaFold. \\u003cem\\u003eNature 2021 596:7873\\u003c/em\\u003e, 596, 583\\u0026ndash;589.\\u003c/li\\u003e\\n\\u003cli\\u003eKendrew,J.C. \\u003cem\\u003eet al.\\u003c/em\\u003e (1958) A Three-Dimensional Model of the Myoglobin Molecule Obtained by X-Ray Analysis. \\u003cem\\u003eNature 1958 181:4610\\u003c/em\\u003e, 181, 662\\u0026ndash;666.\\u003c/li\\u003e\\n\\u003cli\\u003eLaine E, \\u003cem\\u003eet al\\u003c/em\\u003e. (2021) Protein sequence-to-structure learning: Is this the end (-to-end revolution)?. Proteins. 89(12):1770-1786. \\u003c/li\\u003e\\n\\u003cli\\u003eLander,E.S. (1996) The New Genomics: Global Views of Biology. \\u003cem\\u003eScience (1979)\\u003c/em\\u003e, 274, 536\\u0026ndash;539.\\u003c/li\\u003e\\n\\u003cli\\u003eO\\u0026rsquo;Leary,N.A. \\u003cem\\u003eet al.\\u003c/em\\u003e (2016) Reference sequence (RefSeq) database at NCBI: current status, taxonomic expansion, and functional annotation. \\u003cem\\u003eNucleic Acids Res\\u003c/em\\u003e, 44, D733\\u0026ndash;D745.\\u003c/li\\u003e\\n\\u003cli\\u003eSayers,E.W. \\u003cem\\u003eet al.\\u003c/em\\u003e (2024) Database resources of the National Center for Biotechnology Information. \\u003cem\\u003eNucleic Acids Res\\u003c/em\\u003e, 52, D33\\u0026ndash;D43.\\u003c/li\\u003e\\n\\u003cli\\u003eStandley,D.M. \\u003cem\\u003eet al.\\u003c/em\\u003e (2022) The evolution of structural genomics. \\u003cem\\u003eBiophys Rev\\u003c/em\\u003e, 14, 1247.\\u003c/li\\u003e\\n\\u003c/ol\\u003e\"}],\"fulltextSource\":\"\",\"fullText\":\"\",\"funders\":[],\"hasAdminPriorityOnWorkflow\":false,\"hasManuscriptDocX\":true,\"hasOptedInToPreprint\":true,\"hasPassedJournalQc\":\"\",\"hasAnyPriority\":true,\"hideJournal\":false,\"highlight\":\"\",\"institution\":\"\",\"isAcceptedByJournal\":true,\"isAuthorSuppliedPdf\":false,\"isDeskRejected\":\"\",\"isHiddenFromSearch\":false,\"isInQc\":false,\"isInWorkflow\":false,\"isPdf\":false,\"isPdfUpToDate\":true,\"isWithdrawnOrRetracted\":false,\"journal\":{\"display\":true,\"email\":\"info@researchsquare.com\",\"identity\":\"bmc-bioinformatics\",\"isNatureJournal\":false,\"hasQc\":true,\"allowDirectSubmit\":false,\"externalIdentity\":\"binf\",\"sideBox\":\"Learn more about [BMC Bioinformatics](http://bmcbioinformatics.biomedcentral.com/)\",\"snPcode\":\"\",\"submissionUrl\":\"https://www.editorialmanager.com/binf\",\"title\":\"BMC Bioinformatics\",\"twitterHandle\":\"@BMC_Bioinformatics\",\"acdcEnabled\":true,\"dfaEnabled\":false,\"editorialSystem\":\"em\",\"reportingPortfolio\":\"BMC Series\",\"inReviewEnabled\":true,\"inReviewRevisionsEnabled\":true},\"keywords\":\"protein structure, NCBI, UniProt, open access, AlphaFoldDB, Structural genomics, Computational biology, Bioinformatics, API toolkit\",\"lastPublishedDoi\":\"10.21203/rs.3.rs-6583051/v1\",\"lastPublishedDoiUrl\":\"https://doi.org/10.21203/rs.3.rs-6583051/v1\",\"license\":{\"name\":\"CC BY 4.0\",\"url\":\"https://creativecommons.org/licenses/by/4.0/\"},\"manuscriptAbstract\":\"The AlphaFoldDB Structure Extractor is an open-access web server and API toolkit designed to facilitate the bulk download of predicted protein structures from the AlphaFold Database using well-known accession numbers. Addressing the current limitations in extracting structures beyond a restricted list of model organisms and a threshold number, this tool accepts diverse input identifiers, such as NCBI Taxonomy ID, RefSeq accessions, locus tags, and UniProt or AlphaFold accessions, and maps them to UniProt and AlphaFold IDs for structure retrieval. Users can download structure files in PDB, mmCIF, bCIF, or PAE JSON formats. The tool also generates an accompanying ID mapping file to trace input identifiers back to standard accession numbers and reports unmapped IDs separately. An API methodology is also provided for programmatic access, enabling integration into bioinformatics pipelines. AlphaFoldDB Structure Extractor streamlines the structure procurement process from AlphaFold database, empowering researchers in structural and functional genomics with minimal computational expertise.\\nAvailability and implementation: AlphaFoldDB Structure Extractor is freely available as a webserver at https://project.iith.ac.in/sharmaglab/alphafoldextractor/ for downloading structure coordinate files based on the user-provided input of protein accessions. The tool can be accessed programmatically using the API service we have provided.\",\"manuscriptTitle\":\"AlphaFold Database Structure Extractor: A web server and API to download AlphaFold structures using common protein accessions\",\"msid\":\"\",\"msnumber\":\"\",\"nonDraftVersions\":[{\"code\":1,\"date\":\"2025-05-06 14:52:28\",\"doi\":\"10.21203/rs.3.rs-6583051/v1\",\"editorialEvents\":[{\"type\":\"communityComments\",\"content\":0},{\"type\":\"decision\",\"content\":\"Revision requested\",\"date\":\"2025-05-07T15:19:34+00:00\",\"index\":\"\",\"fulltext\":\"\"},{\"type\":\"editorAssigned\",\"content\":\"\",\"date\":\"2025-05-05T08:32:29+00:00\",\"index\":\"\",\"fulltext\":\"\"},{\"type\":\"checksComplete\",\"content\":\"\",\"date\":\"2025-05-05T08:28:09+00:00\",\"index\":\"\",\"fulltext\":\"\"},{\"type\":\"submitted\",\"content\":\"BMC Bioinformatics\",\"date\":\"2025-05-03T08:45:32+00:00\",\"index\":\"\",\"fulltext\":\"\"}],\"status\":\"published\",\"journal\":{\"display\":true,\"email\":\"info@researchsquare.com\",\"identity\":\"bmc-bioinformatics\",\"isNatureJournal\":false,\"hasQc\":true,\"allowDirectSubmit\":false,\"externalIdentity\":\"binf\",\"sideBox\":\"Learn more about [BMC Bioinformatics](http://bmcbioinformatics.biomedcentral.com/)\",\"snPcode\":\"\",\"submissionUrl\":\"https://www.editorialmanager.com/binf\",\"title\":\"BMC Bioinformatics\",\"twitterHandle\":\"@BMC_Bioinformatics\",\"acdcEnabled\":true,\"dfaEnabled\":false,\"editorialSystem\":\"em\",\"reportingPortfolio\":\"BMC Series\",\"inReviewEnabled\":true,\"inReviewRevisionsEnabled\":true}}],\"origin\":\"\",\"ownerIdentity\":\"7eae9e17-7935-49c1-ad95-a99f03f21bb6\",\"owner\":[],\"postedDate\":\"May 6th, 2025\",\"published\":true,\"recentEditorialEvents\":[],\"rejectedJournal\":[],\"revision\":\"\",\"amendment\":\"\",\"status\":\"published-in-journal\",\"subjectAreas\":[],\"tags\":[],\"updatedAt\":\"2025-12-01T15:59:30+00:00\",\"versionOfRecord\":{\"articleIdentity\":\"rs-6583051\",\"link\":\"https://doi.org/10.1186/s12859-025-06303-0\",\"journal\":{\"identity\":\"bmc-bioinformatics\",\"isVorOnly\":false,\"title\":\"BMC Bioinformatics\"},\"publishedOn\":\"2025-11-25 15:57:00\",\"publishedOnDateReadable\":\"November 25th, 2025\"},\"versionCreatedAt\":\"2025-05-06 14:52:28\",\"video\":\"\",\"vorDoi\":\"10.1186/s12859-025-06303-0\",\"vorDoiUrl\":\"https://doi.org/10.1186/s12859-025-06303-0\",\"workflowStages\":[]},\"version\":\"v1\",\"identity\":\"rs-6583051\",\"journalConfig\":\"researchsquare\"},\"__N_SSP\":true},\"page\":\"/article/[identity]/[[...version]]\",\"query\":{\"redirect\":\"/article/rs-6583051\",\"identity\":\"rs-6583051\",\"version\":[\"v1\"]},\"buildId\":\"XKTyCvWXoU3ODBz1xrDgd\",\"isFallback\":false,\"isExperimentalCompile\":false,\"dynamicIds\":[84888],\"gssp\":true,\"scriptLoader\":[]}","source_license":"CC-BY-4.0","license_restricted":false}