{"status":"ok","message-type":"work","message-version":"1.0.0","message":{"indexed":{"date-parts":[[2026,9,2]],"date-time":"2026-09-02T00:16:34Z","timestamp":1788308194295,"version":"build-2803163510"},"reference-count":35,"publisher":"Oxford University Press (OUP)","issue":"9","license":[{"start":{"date-parts":[[2026,8,19]],"date-time":"2026-08-19T00:00:00Z","timestamp":1787097600000},"content-version":"vor","delay-in-days":0,"URL":"https:\/\/creativecommons.org\/licenses\/by\/4.0\/"}],"funder":[{"name":"Amii and New Harvest, Ali Parsaee, Miranda Stahn, and David Staszak"}],"content-domain":{"domain":[],"crossmark-restriction":false},"short-container-title":[],"published-print":{"date-parts":[[2026,8,31]]},"abstract":"<jats:title>Abstract<\/jats:title>\n                  <jats:sec>\n                    <jats:title>Motivation<\/jats:title>\n                    <jats:p>Optimizing cell-culture media for cultivated meat and cellular agriculture is both experimentally demanding and expensive. Traditional approaches use Design of Experiments (DOE) to iterate through candidate formulations until a satisfactory medium emerges. Data-driven methods, and in particular Bayesian optimization (BO), can reach higher objective function values (here, titer) with fewer evaluations than classical DOE. Prior comparisons of optimizers on cell-media data have, however, relied largely on computational simulations rather than measured formulations, leaving researchers without evidence-based guidance on which strategy to adopt.<\/jats:p>\n                  <\/jats:sec>\n                  <jats:sec>\n                    <jats:title>Results<\/jats:title>\n                    <jats:p>We introduce two complementary benchmark protocols and apply them to six published cell-culture datasets. In the \u201cHide-the-Label\u201d protocol each optimizer begins with only a partial view of the data and sequentially reveals hidden formulations, singly or in batches, reproducing the stepwise acquisition of costly laboratory measurements; the \u201cOpen Race\u201d instead measures search efficiency of a method under a fixed evaluation budget. Both protocols were run across four generative surrogate classes [Gaussian process (GP), random forest, neural network, and Bayesian neural network], two difficulty regimes, and batch sizes of one to 20. GP-based BO was consistently the most sample-efficient strategy, locating hidden targets in substantially fewer evaluations than classical DOE designs or random search; its advantage narrowed but remained under injected noise and multi-modality, and optimizers built on more complex surrogates (deeper networks or adaptive ensembles) performed worse rather than better on these harder landscapes. We detail the mathematical basis of the benchmark, criteria for selecting suitable datasets, and how to obtain statistically robust conclusions in the presence of measurement bias and heterogeneity, and we provide a roadmap for selecting an optimization strategy in practice.<\/jats:p>\n                  <\/jats:sec>\n                  <jats:sec>\n                    <jats:title>Availability and implementation<\/jats:title>\n                    <jats:p>Source code and the benchmark datasets are freely available at https:\/\/github.com\/Amii-Applied-AI\/amii-cell-ag-tools\/tree\/main\/active-learning-for-cell-media and archived at https:\/\/doi.org\/10.5281\/zenodo.21501659.<\/jats:p>\n                  <\/jats:sec>","DOI":"10.1093\/bioinformatics\/btag607","type":"journal-article","created":{"date-parts":[[2026,8,18]],"date-time":"2026-08-18T11:39:12Z","timestamp":1787053152000},"source":"Crossref","is-referenced-by-count":0,"title":["A data-driven evaluation of optimization techniques in cell culture media"],"prefix":"10.1093","volume":"42","author":[{"ORCID":"https:\/\/orcid.org\/0009-0003-9576-5806","authenticated-orcid":false,"given":"Ali","family":"Parsaee","sequence":"first","affiliation":[{"name":"Alberta Machine Intelligence Institute (AMII) , Edmonton, AB T6E 4S6,","place":["Canada"]}],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Miranda","family":"Stahn","sequence":"additional","affiliation":[{"name":"New Harvest , Edmonton, AB T6E 1M4,","place":["Canada"]}],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Arian","family":"Amirvaresi","sequence":"additional","affiliation":[{"name":"Department of Food Science and Technology, Texas A&M University , College Station, TX 77843-2256,","place":["United States"]}],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Reza","family":"Ovissipour","sequence":"additional","affiliation":[{"name":"Department of Food Science and Technology, Texas A&M University , College Station, TX 77843-2256,","place":["United States"]}],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"David","family":"Staszak","sequence":"additional","affiliation":[{"name":"Alberta Machine Intelligence Institute (AMII) , Edmonton, AB T6E 4S6,","place":["Canada"]}],"role":[{"vocabulary":"crossref","role":"author"}]}],"member":"286","published-online":{"date-parts":[[2026,8,19]]},"reference":[{"key":"2026090119565380300_btag607-B1","author":"American Pharmaceutical Review","year":"2019"},{"key":"2026090119565380300_btag607-B2","doi-asserted-by":"publisher","first-page":"411","DOI":"10.1097\/01.psy.0000127692.23278.a9","article-title":"What you see may not be what you get: a brief, nontechnical introduction to overfitting in regression-type models","volume":"66","author":"Babyak","year":"2004","journal-title":"Psychosom Med"},{"key":"2026090119565380300_btag607-B3","doi-asserted-by":"publisher","first-page":"1096","DOI":"10.1021\/acs.jcim.8b00839","article-title":"Guacamol: benchmarking models for de novo molecular design","volume":"59","author":"Brown","year":"2019","journal-title":"J Chem Inf Model"},{"key":"2026090119565380300_btag607-B4","author":"Busetto","year":"2025"},{"key":"2026090119565380300_btag607-B5","doi-asserted-by":"publisher","first-page":"2447","DOI":"10.1002\/bit.28132","article-title":"Multi-information source Bayesian optimization of culture media for cellular agriculture","volume":"119","author":"Cosenza","year":"2022","journal-title":"Biotechnol Bioeng"},{"key":"2026090119565380300_btag607-B6","doi-asserted-by":"publisher","first-page":"e2100228","DOI":"10.1002\/biot.202100228","article-title":"Optimization of muscle cell culture media using nonlinear design of experiments","volume":"16","author":"Cosenza","year":"2021","journal-title":"Biotechnol J"},{"key":"2026090119565380300_btag607-B7","doi-asserted-by":"publisher","first-page":"e2300005","DOI":"10.1002\/elsc.202300005","article-title":"Multi-objective Bayesian algorithm automatically discovers low-cost high-growth serum-free media for cellular agriculture application","volume":"23","author":"Cosenza","year":"2023","journal-title":"Eng Life Sci"},{"key":"2026090119565380300_btag607-B8","author":"Frazier","year":"2018"},{"key":"2026090119565380300_btag607-B9","doi-asserted-by":"publisher","first-page":"42","DOI":"10.1016\/j.coche.2018.08.004","article-title":"Media formulation optimization: current and future opportunities","volume":"22","author":"Galbraith","year":"2018","journal-title":"Curr Opin Chem Eng"},{"key":"2026090119565380300_btag607-B10","doi-asserted-by":"publisher","first-page":"308","DOI":"10.1007\/s00253-024-13147-w","article-title":"Explainable AI for CHO cell culture media optimization and prediction of critical quality attribute","volume":"108","author":"Gangwar","year":"2024","journal-title":"Appl Microbiol Biotechnol"},{"key":"2026090119565380300_btag607-B11","doi-asserted-by":"publisher","first-page":"20","DOI":"10.1038\/s41540-023-00284-7","article-title":"Employing active learning in the optimization of culture medium for mammalian cells","volume":"9","author":"Hashizume","year":"2023","journal-title":"NPJ Syst Biol Appl"},{"key":"2026090119565380300_btag607-B12","doi-asserted-by":"publisher","first-page":"108293","DOI":"10.1016\/j.biotechadv.2023.108293","article-title":"Challenges in developing cell culture media using machine learning","volume":"70","author":"Hashizume","year":"2024","journal-title":"Biotechnol Adv"},{"key":"2026090119565380300_btag607-B13","doi-asserted-by":"publisher","first-page":"141","DOI":"10.1016\/j.nbt.2025.07.006","article-title":"Biology-aware machine learning for culture medium optimization","volume":"89","author":"Hashizume","year":"2025","journal-title":"N Biotechnol"},{"key":"2026090119565380300_btag607-B14","doi-asserted-by":"publisher","DOI":"10.26434\/chemrxiv-2023-74w8d"},{"key":"2026090119565380300_btag607-B15","doi-asserted-by":"publisher","first-page":"623","DOI":"10.1007\/s11626-019-00354-8","article-title":"Serum-free media for animal cell culture: advances and applications","volume":"55","author":"Jayme","year":"2019","journal-title":"In Vitro Cell Dev Biol Animal"},{"key":"2026090119565380300_btag607-B16","doi-asserted-by":"publisher","first-page":"144","DOI":"10.1016\/j.ejpb.2021.06.004","article-title":"Design of experiments and design space approaches in the pharmaceutical bioprocess optimization","volume":"166","author":"Kasemiire","year":"2021","journal-title":"Eur J Pharm Biopharm"},{"key":"2026090119565380300_btag607-B17","doi-asserted-by":"publisher","first-page":"108480","DOI":"10.1016\/j.biotechadv.2024.108480","article-title":"Recent advances in culture medium design for enhanced production of monoclonal antibodies in CHO cells: a comparative study of machine learning and systems biology approaches","volume":"78","author":"Kavoni","year":"2025","journal-title":"Biotechnol Adv"},{"key":"2026090119565380300_btag607-B18","doi-asserted-by":"publisher","first-page":"48","DOI":"10.1038\/s42003-019-0296-7","article-title":"On-demand serum-free media formulations for human hematopoietic cell expansion using a high dimensional search algorithm","volume":"2","author":"Kim","year":"2019","journal-title":"Commun Biol"},{"key":"2026090119565380300_btag607-B19","doi-asserted-by":"publisher","first-page":"1191","DOI":"10.1002\/btpr.67","article-title":"Bioprocess optimization using design-of-experiments methodology","volume":"24","author":"Mandenius","year":"2008","journal-title":"Biotechnol Prog"},{"key":"2026090119565380300_btag607-B20","volume-title":"Design and Analysis of Experiments","author":"Montgomery","year":"2017"},{"key":"2026090119565380300_btag607-B21","author":"Myers"},{"key":"2026090119565380300_btag607-B22","doi-asserted-by":"publisher","first-page":"6055","DOI":"10.1038\/s41467-025-61113-5","article-title":"Accelerating cell culture media development using Bayesian optimization-based iterative experimental design","volume":"16","author":"Narayanan","year":"2025","journal-title":"Nat Commun"},{"key":"2026090119565380300_btag607-B23","first-page":"1799","article-title":"Design of experiments for screening in industrial biotechnology: a review","volume":"116","author":"O\u2019Kennedy","year":"2019","journal-title":"Biotechnol Bioeng"},{"key":"2026090119565380300_btag607-B24","doi-asserted-by":"publisher","first-page":"109591","DOI":"10.1016\/j.bej.2024.109591","article-title":"A data-driven approach for cell culture medium optimization","volume":"214","author":"Ozawa","year":"2025","journal-title":"Biochem Eng J"},{"key":"2026090119565380300_btag607-B25","doi-asserted-by":"publisher","first-page":"57","DOI":"10.1007\/s10616-007-9074-3","article-title":"Development of serum-free media in cho-dg44 cells using a central composite statistical design","volume":"54","author":"Parampalli","year":"2007","journal-title":"Cytotechnology"},{"key":"2026090119565380300_btag607-B26","first-page":"108031","article-title":"Cultivated meat from stem cells: challenges and prospects","volume":"162","author":"Post","year":"2020","journal-title":"Meat Sci"},{"key":"2026090119565380300_btag607-B27","doi-asserted-by":"publisher","first-page":"315","DOI":"10.1002\/elsc.201800168","article-title":"A statistical approach to improve compound screening in cell culture media","volume":"19","author":"Puente-Massaguer","year":"2019","journal-title":"Eng Life Sci"},{"key":"2026090119565380300_btag607-B28","volume-title":"Gaussian Processes for Machine Learning","author":"Rasmussen","year":"2006"},{"key":"2026090119565380300_btag607-B29","doi-asserted-by":"publisher","first-page":"508","DOI":"10.3390\/pr9090508","article-title":"Model-based design of experiments for high-dimensional inputs supported by machine-learning methods","volume":"9","author":"Seufert","year":"2021","journal-title":"Processes"},{"key":"2026090119565380300_btag607-B30","doi-asserted-by":"publisher","first-page":"148","DOI":"10.1109\/JPROC.2015.2494218","article-title":"Taking the human out of the loop: a review of Bayesian optimization","volume":"104","author":"Shahriari","year":"2016","journal-title":"Proc IEEE"},{"key":"2026090119565380300_btag607-B31"},{"key":"2026090119565380300_btag607-B32","doi-asserted-by":"publisher","DOI":"10.1007\/978-1-4612-1494-6","volume-title":"Interpolation of Spatial Data: Some Theory for Kriging","author":"Stein","year":"1999"},{"key":"2026090119565380300_btag607-B33","article-title":"Serum-free media for cultivated meat production","author":"Swartz","year":"2021"},{"key":"2026090119565380300_btag607-B34","doi-asserted-by":"publisher","first-page":"1424012","DOI":"10.3389\/frai.2024.1424012","article-title":"Artificial intelligence and machine learning applications for cultured meat","volume":"7","author":"Todhunter","year":"2024","journal-title":"Front Artif Intell"},{"key":"2026090119565380300_btag607-B35","doi-asserted-by":"publisher","first-page":"1195294","DOI":"10.3389\/fbioe.2023.1195294","article-title":"A review of algorithmic approaches for cell culture media optimization","volume":"11","author":"Zhou","year":"2023","journal-title":"Front Bioeng Biotechnol"}],"container-title":["Bioinformatics"],"original-title":[],"language":"en","link":[{"URL":"https:\/\/academic.oup.com\/bioinformatics\/advance-article-pdf\/doi\/10.1093\/bioinformatics\/btag607\/70708795\/btag607.pdf","content-type":"application\/pdf","content-version":"am","intended-application":"syndication"},{"URL":"https:\/\/academic.oup.com\/bioinformatics\/article-pdf\/42\/9\/btag607\/70708795\/btag607.pdf","content-type":"application\/pdf","content-version":"vor","intended-application":"syndication"},{"URL":"https:\/\/academic.oup.com\/bioinformatics\/article-pdf\/42\/9\/btag607\/70708795\/btag607.pdf","content-type":"unspecified","content-version":"vor","intended-application":"similarity-checking"}],"deposited":{"date-parts":[[2026,9,1]],"date-time":"2026-09-01T23:57:03Z","timestamp":1788307023000},"score":1,"resource":{"primary":{"URL":"https:\/\/academic.oup.com\/bioinformatics\/article\/doi\/10.1093\/bioinformatics\/btag607\/8766006"}},"subtitle":[],"editor":[{"given":"Lenore","family":"Cowen","sequence":"additional","affiliation":[],"role":[{"vocabulary":"crossref","role":"editor"}]}],"short-title":[],"issued":{"date-parts":[[2026,8,19]]},"references-count":35,"journal-issue":{"issue":"9","published-print":{"date-parts":[[2026,8,31]]}},"URL":"https:\/\/doi.org\/10.1093\/bioinformatics\/btag607","relation":{},"ISSN":["1367-4803","1367-4811"],"issn-type":[{"value":"1367-4803","type":"print"},{"value":"1367-4811","type":"electronic"}],"subject":[],"published-other":{"date-parts":[[2026,9]]},"published":{"date-parts":[[2026,8,19]]},"article-number":"btag607"}}