TadA-Bench: A Million-Variant Benchmark for Future-Round Discovery Toward Agentic Protein Engineering

Jin Gao, Juntu Zhao, Zirui Zeng, Jiaqi Shen, Junhao Shi, Dukun Zhao, Yuming Lu, Dequan Wang
Proceedings of the 43rd International Conference on Machine Learning, PMLR 306:33997-34014, 2026.

Abstract

AI for scientific discovery is entering an agentic era, where protein-engineering systems are expected to prioritize future wet-lab experiments rather than merely fit static measurements. We introduce TadA-Bench, a million-variant wet-lab replay benchmark from 31 TadA directed-evolution rounds for future-round discovery toward agentic protein engineering. TadA-Bench preserves the campaign chronology and defines a fixed-data replay task: given earlier experimental rounds, models rank variants that appear only in later rounds. It provides aligned DNA, RNA, and protein views, and uses Seq2Graph, a graph-based label-unification pipeline, to reconcile noisy enrichment measurements into consistent cross-round activity labels. Random-split controls show strong interpolation, but future-round ranking and finite-budget candidate selection are much weaker. Controlled analyses suggest that evolutionary coverage is more informative than local data density, positioning TadA-Bench as a reproducible wet-lab replay substrate for future-round discovery toward agentic protein engineering; the data and code are released on Hugging Face and GitHub.

Cite this Paper

BibTeX
@InProceedings{pmlr-v306-gao26ao, title = {{T}ad{A}-Bench: A Million-Variant Benchmark for Future-Round Discovery Toward Agentic Protein Engineering}, author = {Gao, Jin and Zhao, Juntu and Zeng, Zirui and Shen, Jiaqi and Shi, Junhao and Zhao, Dukun and Lu, Yuming and Wang, Dequan}, booktitle = {Proceedings of the 43rd International Conference on Machine Learning}, pages = {33997--34014}, year = {2026}, editor = {Zhang, Tong and Dudik, Miroslav and Jaggi, Martin and Agarwal, Alekh and Li, Sharon and Schuurmans, Dale and Zhu, Jerry and Berkenkamp, Felix and Dong, Hanze and Bietti, Alberto}, volume = {306}, series = {Proceedings of Machine Learning Research}, month = {06--11 Jul}, publisher = {PMLR}, pdf = {https://raw.githubusercontent.com/mlresearch/v306/main/assets/gao26ao/gao26ao.pdf}, url = {https://proceedings.mlr.press/v306/gao26ao.html}, abstract = {AI for scientific discovery is entering an agentic era, where protein-engineering systems are expected to prioritize future wet-lab experiments rather than merely fit static measurements. We introduce TadA-Bench, a million-variant wet-lab replay benchmark from 31 TadA directed-evolution rounds for future-round discovery toward agentic protein engineering. TadA-Bench preserves the campaign chronology and defines a fixed-data replay task: given earlier experimental rounds, models rank variants that appear only in later rounds. It provides aligned DNA, RNA, and protein views, and uses Seq2Graph, a graph-based label-unification pipeline, to reconcile noisy enrichment measurements into consistent cross-round activity labels. Random-split controls show strong interpolation, but future-round ranking and finite-budget candidate selection are much weaker. Controlled analyses suggest that evolutionary coverage is more informative than local data density, positioning TadA-Bench as a reproducible wet-lab replay substrate for future-round discovery toward agentic protein engineering; the data and code are released on Hugging Face and GitHub.} }
Endnote
%0 Conference Paper %T TadA-Bench: A Million-Variant Benchmark for Future-Round Discovery Toward Agentic Protein Engineering %A Jin Gao %A Juntu Zhao %A Zirui Zeng %A Jiaqi Shen %A Junhao Shi %A Dukun Zhao %A Yuming Lu %A Dequan Wang %B Proceedings of the 43rd International Conference on Machine Learning %C Proceedings of Machine Learning Research %D 2026 %E Tong Zhang %E Miroslav Dudik %E Martin Jaggi %E Alekh Agarwal %E Sharon Li %E Dale Schuurmans %E Jerry Zhu %E Felix Berkenkamp %E Hanze Dong %E Alberto Bietti %F pmlr-v306-gao26ao %I PMLR %P 33997--34014 %U https://proceedings.mlr.press/v306/gao26ao.html %V 306 %X AI for scientific discovery is entering an agentic era, where protein-engineering systems are expected to prioritize future wet-lab experiments rather than merely fit static measurements. We introduce TadA-Bench, a million-variant wet-lab replay benchmark from 31 TadA directed-evolution rounds for future-round discovery toward agentic protein engineering. TadA-Bench preserves the campaign chronology and defines a fixed-data replay task: given earlier experimental rounds, models rank variants that appear only in later rounds. It provides aligned DNA, RNA, and protein views, and uses Seq2Graph, a graph-based label-unification pipeline, to reconcile noisy enrichment measurements into consistent cross-round activity labels. Random-split controls show strong interpolation, but future-round ranking and finite-budget candidate selection are much weaker. Controlled analyses suggest that evolutionary coverage is more informative than local data density, positioning TadA-Bench as a reproducible wet-lab replay substrate for future-round discovery toward agentic protein engineering; the data and code are released on Hugging Face and GitHub.
APA
Gao, J., Zhao, J., Zeng, Z., Shen, J., Shi, J., Zhao, D., Lu, Y. & Wang, D.. (2026). TadA-Bench: A Million-Variant Benchmark for Future-Round Discovery Toward Agentic Protein Engineering. Proceedings of the 43rd International Conference on Machine Learning, in Proceedings of Machine Learning Research 306:33997-34014 Available from https://proceedings.mlr.press/v306/gao26ao.html.

Related Material