{"status":"ok","message-type":"work","message-version":"1.0.0","message":{"institution":[{"name":"bioRxiv"}],"indexed":{"date-parts":[[2026,10,9]],"date-time":"2026-10-09T03:48:56Z","timestamp":1791517736342,"version":"4.3.4"},"posted":{"date-parts":[[2026,9,23]]},"group-title":"Bioengineering","reference-count":69,"publisher":"openRxiv","license":[{"start":{"date-parts":[[2026,9,23]],"date-time":"2026-09-23T00:00:00Z","timestamp":1790121600000},"content-version":"vor","delay-in-days":0,"URL":"http:\/\/creativecommons.org\/licenses\/by\/4.0\/"}],"content-domain":{"domain":[],"crossmark-restriction":false},"short-container-title":[],"accepted":{"date-parts":[[2026,10,5]]},"abstract":"<jats:title>Abstract<\/jats:title>\n                <jats:p>\n                  Progress in artificial intelligence has produced a rapidly growing ecosystem of methods for\n                  <jats:italic>de novo<\/jats:italic>\n                  protein design. With access to many specialized and often complementary tools, how does one use them effectively, especially with a finite compute budget? Here, we introduce\n                  <jats:bold>Target-adaptive Rescue\u2013Explore\u2013eXploit (T-REX)<\/jats:bold>\n                  , an agentic protein binder design framework that orchestrates multiple state-of-the-art protein generative models and structure evaluators. Over the course of a design campaign, T-REX leverages large language model (LLM) agents to reason over accumulating outcomes and decide whether to rescue promising candidates, explore alternative settings, or exploit productive routes, while a deterministic controller validates and schedules proposed actions. Across seven targets, we find that T-REX adaptively allocates its compute across methods in a target-dependent manner and achieves the highest throughput of structurally distinct hits compared to all baselines. These results suggest that adaptive orchestration can complement increasingly powerful protein design models, and we openly release T-REX to support compute-efficient binder design workflows.\n                <\/jats:p>","DOI":"10.64898\/2026.09.22.753604","type":"posted-content","created":{"date-parts":[[2026,9,23]],"date-time":"2026-09-23T17:50:22Z","timestamp":1790185822000},"source":"Crossref","is-referenced-by-count":0,"title":["Agentic campaign control for high-throughput de novo binder design"],"prefix":"10.64898","author":[{"ORCID":"https:\/\/orcid.org\/0000-0003-0572-6065","authenticated-orcid":false,"given":"Minkyu","family":"Jeon","sequence":"first","affiliation":[{"name":"Department of Computer Science, Princeton University"}],"role":[{"vocabulary":"crossref","role":"author"}]},{"ORCID":"https:\/\/orcid.org\/0000-0002-4113-5185","authenticated-orcid":false,"given":"Jinyeop","family":"Song","sequence":"additional","affiliation":[{"name":"Department of Physics, Massachusetts Institute of Technology"}],"role":[{"vocabulary":"crossref","role":"author"}]},{"ORCID":"https:\/\/orcid.org\/0009-0009-6898-7933","authenticated-orcid":false,"given":"Jina","family":"Kim","sequence":"additional","affiliation":[{"name":"Department of Computer Science, Princeton University"}],"role":[{"vocabulary":"crossref","role":"author"}]},{"ORCID":"https:\/\/orcid.org\/0000-0001-6345-1907","authenticated-orcid":false,"given":"Ellen D.","family":"Zhong","sequence":"additional","affiliation":[{"name":"Department of Computer Science, Princeton University"}],"role":[{"vocabulary":"crossref","role":"author"}]}],"member":"54368","reference":[{"key":"2026100819500477000_2026.09.22.753604v3.1","doi-asserted-by":"crossref","unstructured":"J. Abramson , J. Adler , J. Dunger , R. Evans , T. Green , A. Pritzel , O. Ronneberger , L. Willmore , A. J. Ballard , J. Bambrick , S. W. Bodenstein , D. A. Evans , C.-C. Hung , M. O\u2019Neill , D. Reiman , K. Tunya-suvunakool , Z. Wu , A. \u017demgulyte, E. Arvaniti, C. Beattie , O. Bertolli , A. Bridgland , A. Cherepanov , M. Congreve , A. I. Cowen-Rivers , A. Cowie , M. Figurnov , F. B. Fuchs , H. Gladman , R. Jain , Y. A. Khan , C. M. R. Low , K. Perlin , A. Potapenko , P. Savy , S. Singh , A. Stecula , A. Thillaisundaram , C. Tong , S. Yakneen , E. D. Zhong , M. Zielinski , A. \u017d\u00eddek , V. Bapst , P. Kohli , M. Jaderberg , D. Hassabis , and J. M. Jumper . Accurate structure prediction of biomolecular interactions with AlphaFold 3. Nature, May 2024.","DOI":"10.1038\/s41586-024-08416-7"},{"key":"2026100819500477000_2026.09.22.753604v3.2","doi-asserted-by":"crossref","unstructured":"A. Alsaadi , J. Ash , M. Titov , M. Turilli , A. Merzky , S. Jha , and S. Khare . Adaptive protein design protocols and middleware. In 2025 IEEE International Parallel and Distributed Processing Symposium Workshops (IPDPSW), pages 1011\u20131015. IEEE, 2025.","DOI":"10.1109\/IPDPSW66978.2025.00157"},{"key":"2026100819500477000_2026.09.22.753604v3.3","unstructured":"E. Ayg\u00fcn , A. Belyaeva , G. Comanici , M. Coram , H. Cui , J. Garrison , R. Johnston , A. Kast , C. Y. McLean , P. Norgaard , et al. An ai system to help scientists write expert-level empirical software. Nature, pages 1\u20133, 2026."},{"issue":"8","key":"2026100819500477000_2026.09.22.753604v3.4","doi-asserted-by":"crossref","first-page":"btae492","DOI":"10.1093\/bioinformatics\/btae492","article-title":"Locuaz: an in silico platform for protein binders optimization","volume":"40","year":"2024","journal-title":"Bioinformatics"},{"issue":"1","key":"2026100819500477000_2026.09.22.753604v3.5","first-page":"2023","article-title":"Improving de novo protein binder design with deep learning","volume":"14","year":"2625","journal-title":"Nature Communications"},{"key":"2026100819500477000_2026.09.22.753604v3.6","doi-asserted-by":"publisher","DOI":"10.1038\/s41586-023-06792-0"},{"key":"2026100819500477000_2026.09.22.753604v3.7","doi-asserted-by":"crossref","unstructured":"S. Candido , T. Hayes , A. Derry , R. Rao , Z. Lin , R. Verkuil , B. Z. Wu , J. S. Lee , E. S. Bruguera , J. A. Keval , et al. Language modeling materializes a world model of protein biology. bioRxiv, 2026.","DOI":"10.64898\/2026.06.03.729735"},{"key":"2026100819500477000_2026.09.22.753604v3.8","unstructured":"Chai Discovery , J. Boitreaud , J. Dent , M. McPartlon , J. Meier , V. Reis , A. Rogozhnikov , and K. Wu . Chai-1: Decoding the molecular interactions of life. bioRxiv, page 2024.10.10.615955, Oct. 2024."},{"key":"2026100819500477000_2026.09.22.753604v3.9","doi-asserted-by":"publisher","DOI":"10.1093\/bioinformatics\/btq007"},{"key":"2026100819500477000_2026.09.22.753604v3.10","doi-asserted-by":"publisher","DOI":"10.1038\/nature01392"},{"key":"2026100819500477000_2026.09.22.753604v3.11","doi-asserted-by":"publisher","DOI":"10.1038\/s41587-024-02133-2"},{"key":"2026100819500477000_2026.09.22.753604v3.12","unstructured":"Claude Science and A. Shanehsazzadeh. Autonomous de novo protein binder design with Claude, Aug. 2026. URL https:\/\/www-cdn.anthropic.com\/30bf50e22a01388bb29bf077ee3f244531594b7a.pdf. Anthropic technical report."},{"key":"2026100819500477000_2026.09.22.753604v3.13","doi-asserted-by":"publisher","DOI":"10.1126\/science.add2187"},{"key":"2026100819500477000_2026.09.22.753604v3.14","doi-asserted-by":"crossref","unstructured":"K. Didi , D. Reidenbach , M. Penner , S. Ravi , M. Case , M. Nichols , E. Swanson , A. Reis , M. Prescott , Y. Qian , et al. Latent generative search unlocks de novo design of untapped biomolecular interactions at scale, 2026a.","DOI":"10.64898\/2026.09.12.751118"},{"key":"2026100819500477000_2026.09.22.753604v3.15","unstructured":"K. Didi , Z. Zhang , G. Zhou , D. Reidenbach , Z. Cao , S. Cha , T. Geffner , C. Dallago , J. Tang , M. Bronstein , et al. Scaling atomistic protein binder design with generative pretraining and test-time compute. In International Conference on Learning Representations , volume 2026, pages 61056\u201361109, 2026b."},{"issue":"10","key":"2026100819500477000_2026.09.22.753604v3.16","doi-asserted-by":"crossref","first-page":"962","DOI":"10.1038\/s43588-025-00849-y","article-title":"Scitoolagent: a knowledge-graph-driven scientific agent for multitool integration","volume":"5","year":"2025","journal-title":"Nature Computational Science"},{"key":"2026100819500477000_2026.09.22.753604v3.17","unstructured":"R. L. Dunbrack Jr . Res ipsae loquunt: What\u2019s wrong with alphafold\u2019s iptm score and how to fix it. Biorxiv, 2025."},{"key":"2026100819500477000_2026.09.22.753604v3.18","doi-asserted-by":"crossref","unstructured":"R. Evans , M. O\u2019neill , A. Pritzel , N. Antropova , A. Senior , T. Green , A. \u017d\u00eddek , R. Bates , S. Blackwell , J. Yim , et al. Protein complex prediction with alphafold-multimer. biorxiv, 2021.","DOI":"10.1101\/2021.10.04.463034"},{"key":"2026100819500477000_2026.09.22.753604v3.19","unstructured":"F. Ge , J. Zhu , L. Zhang , H. Xiao , X. Bao , F. Xie , D. Chen , Y. Lu , Y. Wang , Z. Guan , et al. Autobinder agent: An mcp-based agent for end-to-end protein binder design. arXiv preprint arXiv:2602.00019, 2026."},{"issue":"7","key":"2026100819500477000_2026.09.22.753604v3.20","doi-asserted-by":"crossref","first-page":"1389","DOI":"10.1039\/D4DD00013G","article-title":"Protagents: protein discovery via large language model multi-agent collaborations combining physics and machine learning","volume":"3","year":"2024","journal-title":"Digital Discovery"},{"issue":"11","key":"2026100819500477000_2026.09.22.753604v3.21","doi-asserted-by":"crossref","first-page":"e1013747","DOI":"10.1371\/journal.pcbi.1013747","article-title":"Automated and modular protein binder design with binderflow","volume":"21","year":"2025","journal-title":"PLoS computational biology"},{"key":"2026100819500477000_2026.09.22.753604v3.22","doi-asserted-by":"publisher","DOI":"10.1126\/science.ads0018"},{"key":"2026100819500477000_2026.09.22.753604v3.23","doi-asserted-by":"crossref","unstructured":"C. Hsu , R. Verkuil , J. Liu , Z. Lin , B. Hie , T. Sercu , A. Lerer , and A. Rives . Learning inverse folding from millions of predicted structures. In International Conference on Machine Learning, pages 8946\u20138970. PMLR, June 2022.","DOI":"10.1101\/2022.04.10.487779"},{"key":"2026100819500477000_2026.09.22.753604v3.24","unstructured":"J. Ingraham , V. K. Garg , R. Barzilay , and T. Jaakkola . Generative models for graph-based protein design. In Neural Information Processing Systems, 2019."},{"key":"2026100819500477000_2026.09.22.753604v3.25","doi-asserted-by":"publisher","DOI":"10.1038\/s41586-023-06728-8"},{"issue":"9","key":"2026100819500477000_2026.09.22.753604v3.26","doi-asserted-by":"crossref","first-page":"1429","DOI":"10.1038\/s42256-025-01100-z","article-title":"Efficient protein structure generation with sparse denoising models","volume":"7","year":"2025","journal-title":"Nature machine intelligence"},{"key":"2026100819500477000_2026.09.22.753604v3.27","doi-asserted-by":"publisher","DOI":"10.1038\/S41586-021-03819-2"},{"issue":"5","key":"2026100819500477000_2026.09.22.753604v3.28","first-page":"922","article-title":"A solution for the best rotation to relate two sets of vectors","volume":"32","year":"1976","journal-title":"Foundations of Crystallography"},{"issue":"12","key":"2026100819500477000_2026.09.22.753604v3.29","doi-asserted-by":"crossref","first-page":"2577","DOI":"10.1002\/bip.360221211","article-title":"Dictionary of protein secondary structure: pattern recognition of hydrogen-bonded and geometrical features","volume":"22","year":"1983","journal-title":"Biopolymers: Original Research on Biomolecules"},{"key":"2026100819500477000_2026.09.22.753604v3.30","unstructured":"A. Karpathy . autoresearch: AI agents running research on single-GPU nanochat training automatically. GitHub repository, 2026. URL https:\/\/github.com\/karpathy\/autoresearch. Accessed September 15, 2026."},{"key":"2026100819500477000_2026.09.22.753604v3.31","doi-asserted-by":"crossref","unstructured":"W. Kwon , Z. Li , S. Zhuang , Y. Sheng , L. Zheng , C. H. Yu , J. Gonzalez , H. Zhang , and I. Stoica . Efficient memory management for large language model serving with pagedattention. In Proceedings of the 29th symposium on operating systems principles, pages 611\u2013626, 2023.","DOI":"10.1145\/3600006.3613165"},{"key":"2026100819500477000_2026.09.22.753604v3.32","doi-asserted-by":"publisher","DOI":"10.1038\/s41586-020-2180-5"},{"key":"2026100819500477000_2026.09.22.753604v3.33","doi-asserted-by":"publisher","DOI":"10.1126\/science.ade2574"},{"issue":"54","key":"2026100819500477000_2026.09.22.753604v3.34","first-page":"1","article-title":"Smac3: A versatile bayesian optimization package for hyperparameter optimization","volume":"23","year":"2022","journal-title":"Journal of Machine Learning Research"},{"key":"2026100819500477000_2026.09.22.753604v3.35","doi-asserted-by":"publisher","DOI":"10.1038\/s41580-024-00718-y"},{"key":"2026100819500477000_2026.09.22.753604v3.36","unstructured":"C. Liu , M. Ren , J. Guan , C. Gong , J. Sun , X. Chen , and W. Xiao . Protdbench: A unified benchmark of protein binder design and evaluation. arXiv preprint arXiv:2605.04118, 2026."},{"key":"2026100819500477000_2026.09.22.753604v3.37","doi-asserted-by":"crossref","unstructured":"C. Lu , C. Lu , R. T. Lange , Y. Yamada , S. Hu , J. Foerster , D. Ha , and J. Clune . Towards end-to-end automation of ai research. Nature, 651(8107):914\u2013919, 2026.","DOI":"10.1038\/s41586-026-10265-5"},{"issue":"44","key":"2026100819500477000_2026.09.22.753604v3.38","doi-asserted-by":"crossref","first-page":"13732","DOI":"10.1002\/anie.201707707","article-title":"Bioactive macrocyclic inhibitors of the pd-1\/pd-l1 immune checkpoint","volume":"56","year":"2017","journal-title":"Angewandte Chemie International Edition"},{"key":"2026100819500477000_2026.09.22.753604v3.39","doi-asserted-by":"publisher","DOI":"10.1016\/j.str.2008.10.019"},{"key":"2026100819500477000_2026.09.22.753604v3.40","doi-asserted-by":"publisher","DOI":"10.1016\/j.bpj.2015.08.015"},{"key":"2026100819500477000_2026.09.22.753604v3.41","doi-asserted-by":"crossref","unstructured":"A. T. Merchant , D. Guo , B. Viggiano , L. Brennan-Almaraz , E. Hur , T. Mai , P. Yin , S. H. King , E. A. Ashley , and B. L. Hie . A high-level programming language for generative biology with proto. bioRxiv, page 2026.06.22.733870, June 2026.","DOI":"10.64898\/2026.06.22.733870"},{"key":"2026100819500477000_2026.09.22.753604v3.42","unstructured":"S. Ouyang , Z. Dong , Y. Luo , K. Su , H. Zhao , M. Shi , and B. Du . Pdagent: An llm-driven autonomous agent framework towards* in silico* protein design via directed mutation. In Forty-third International Conference on Machine Learning, 2026."},{"key":"2026100819500477000_2026.09.22.753604v3.43","doi-asserted-by":"publisher","DOI":"10.5281\/zenodo.15161108"},{"key":"2026100819500477000_2026.09.22.753604v3.44","doi-asserted-by":"publisher","DOI":"10.1038\/s41586-025-09076-9"},{"key":"2026100819500477000_2026.09.22.753604v3.45","doi-asserted-by":"crossref","unstructured":"S. Passaro , G. Corso , J. Wohlwend , M. Reveiz , S. Thaler , V. R. Somnath , N. Getz , T. Portnoi , J. Roy , H. Stark , et al. Boltz-2: Towards accurate and efficient binding affinity prediction. BioRxiv, 2025.","DOI":"10.1101\/2025.06.14.659707"},{"key":"2026100819500477000_2026.09.22.753604v3.46","doi-asserted-by":"publisher","DOI":"10.1002\/PRO.3943"},{"key":"2026100819500477000_2026.09.22.753604v3.47","unstructured":"Protenix Team , Y. Zhang , C. Gong , H. Zhang , W. Ma , Z. Liu , X. Chen , J. Guan , L. Wang , Y. Yang , Y. Xia , and W. Xiao . Protenix-v1: Toward high-accuracy open-source biomolecular structure prediction. bioRxiv, page 2026.02.05.703733, Feb. 2026."},{"issue":"1","key":"2026100819500477000_2026.09.22.753604v3.48","first-page":"2026","article-title":"Biopipelines: Accessible computational protein and ligand design for chemical biologists","volume":"35","year":"0129","journal-title":"Computational and Structural Biotechnology Journal"},{"issue":"3","key":"2026100819500477000_2026.09.22.753604v3.49","doi-asserted-by":"crossref","first-page":"203","DOI":"10.1007\/s10472-011-9258-6","article-title":"Multi-armed bandits with episode context","volume":"61","year":"2011","journal-title":"Annals of Mathematics and Artificial Intelligence"},{"issue":"2","key":"2026100819500477000_2026.09.22.753604v3.50","doi-asserted-by":"crossref","first-page":"e70464","DOI":"10.1002\/pro.70464","article-title":"Proteindj: A high-performance and modular protein design pipeline","volume":"35","year":"2026","journal-title":"Protein Science"},{"key":"2026100819500477000_2026.09.22.753604v3.51","doi-asserted-by":"publisher","DOI":"10.1038\/nature24270"},{"key":"2026100819500477000_2026.09.22.753604v3.52","first-page":"55810","volume":"267","year":"2025","journal-title":"A general framework for inference-time scaling and steering of diffusion models"},{"key":"2026100819500477000_2026.09.22.753604v3.53","doi-asserted-by":"crossref","unstructured":"H. Stark , F. Faltings , M. Choi , Y. Xie , E. Hur , T. O\u2019Donnell , A. Bushuiev , T. U\u00e7ar , S. Passaro , W. Mao , et al. Boltzgen: Toward universal binder design. BioRxiv, 2025.","DOI":"10.1101\/2025.11.20.689494"},{"key":"2026100819500477000_2026.09.22.753604v3.54","doi-asserted-by":"publisher","DOI":"10.1038\/nbt.3988"},{"key":"2026100819500477000_2026.09.22.753604v3.55","doi-asserted-by":"publisher","DOI":"10.1038\/S41586-025-09442-9"},{"key":"2026100819500477000_2026.09.22.753604v3.56","unstructured":"Q. Team . Qwen3. 6-27b: Flagship-level coding in a 27b dense model, 2026."},{"key":"2026100819500477000_2026.09.22.753604v3.57","unstructured":"J. Teneggi , S. Turzo , T. Marwah , A. Bietti , P. D. Renfrew , V. K. Mulligan , and S. Golkar . Protein design with agent rosetta: A case study for specialized scientific agents. arXiv preprint arXiv:2603.15952, 2026."},{"key":"2026100819500477000_2026.09.22.753604v3.58","doi-asserted-by":"publisher","DOI":"10.1038\/s41587-023-01773-0"},{"issue":"3","key":"2026100819500477000_2026.09.22.753604v3.59","doi-asserted-by":"crossref","first-page":"btaf107","DOI":"10.1093\/bioinformatics\/btaf107","article-title":". actifptm: a refined confidence metric of alphafold2 predictions involving flexible regions","volume":"41","year":"2025","journal-title":"Bioinformatics"},{"key":"2026100819500477000_2026.09.22.753604v3.60","doi-asserted-by":"publisher","DOI":"10.1038\/s41586-023-06953-1"},{"key":"2026100819500477000_2026.09.22.753604v3.61","doi-asserted-by":"crossref","unstructured":"J. Vermorel and M. Mohri . Multi-armed bandit algorithms and empirical evaluation. In European conference on machine learning, pages 437\u2013448. Springer, 2005.","DOI":"10.1007\/11564096_42"},{"issue":"1","key":"2026100819500477000_2026.09.22.753604v3.62","doi-asserted-by":"crossref","first-page":"52","DOI":"10.1177\/10943420241288242","article-title":"Employing artificial intelligence to steer exascale workflows with colmena","volume":"39","year":"2025","journal-title":"The International Journal of High Performance Computing Applications"},{"key":"2026100819500477000_2026.09.22.753604v3.63","doi-asserted-by":"publisher","DOI":"10.1038\/s41586-023-06415-8"},{"key":"2026100819500477000_2026.09.22.753604v3.64","doi-asserted-by":"crossref","unstructured":"J. Wohlwend , G. Corso , S. Passaro , N. Getz , M. Reveiz , K. Leidal , W. Swiderski , L. Atkinson , T. Portnoi , I. Chinn , J. Silterra , T. Jaakkola , and R. Barzilay . Boltz-1 democratizing biomolecular interaction modeling. bioRxivorg, page 2024.11.19.624167, May 2025.","DOI":"10.1101\/2024.11.19.624167"},{"key":"2026100819500477000_2026.09.22.753604v3.65","doi-asserted-by":"publisher","DOI":"10.1093\/bioinformatics\/btq066"},{"key":"2026100819500477000_2026.09.22.753604v3.66","doi-asserted-by":"publisher","DOI":"10.1002\/pro.70547"},{"key":"2026100819500477000_2026.09.22.753604v3.67","doi-asserted-by":"publisher","DOI":"10.1093\/bioinformatics\/btw514"},{"key":"2026100819500477000_2026.09.22.753604v3.68","unstructured":"Y. Yamada , R. T. Lange , C. Lu , S. Hu , C. Lu , J. Foerster , J. Clune , and D. Ha . The AI scientist-v2: Workshop-level automated scientific discovery via agentic tree search. arXiv preprint arXiv:2504.08066, 2025. URL https:\/\/arxiv.org\/abs\/2504.08066."},{"key":"2026100819500477000_2026.09.22.753604v3.69","doi-asserted-by":"publisher","DOI":"10.1038\/s41586-026-10328-7"}],"container-title":[],"original-title":[],"link":[{"URL":"https:\/\/syndication.highwire.org\/content\/doi\/10.64898\/2026.09.22.753604","content-type":"unspecified","content-version":"vor","intended-application":"similarity-checking"}],"deposited":{"date-parts":[[2026,10,9]],"date-time":"2026-10-09T02:50:20Z","timestamp":1791514220000},"score":1,"resource":{"primary":{"URL":"http:\/\/biorxiv.org\/lookup\/doi\/10.64898\/2026.09.22.753604"}},"subtitle":[],"short-title":[],"issued":{"date-parts":[[2026,9,23]]},"references-count":69,"URL":"https:\/\/doi.org\/10.64898\/2026.09.22.753604","relation":{},"subject":[],"published":{"date-parts":[[2026,9,23]]},"subtype":"preprint"}}