Information-Dense Synthesis for Molecular Discovery

Abstract

Machine learning can accelerate molecular discovery by designing molecules and planning experiments. However, many scientific challenges demand molecules with very rare properties, and in this sparse setting, existing algorithms offer little gain over random guessing. We propose a method to efficiently search large regions of molecular space using algorithmically controlled stochastic synthesis. Rather than design, make and test individual molecules, we design and make complex mixtures, test them as a pool, then deconvolute the molecule-activity map. We optimize synthesis to encode maximal information. Theoretically, this approach can reduce the number of experiments required to find the optimal molecule among d candidates from 𝒪(d) to 𝒪(log d) or 𝒪(1). In simulation, on estimated protein fitness landscapes, it finds active molecules with an order of magnitude fewer experiments than existing Bayesian optimization methods.

Special Issue “Molecular Mechanism Discovery of the Bioactive Phytochemicals against Different Diseases Based on Network Pharmacology and Molecular Docking, 2nd Edition”
Baojun Xu
International Journal of Molecular Sciences27 Oct 2026Peer-reviewedCC BY 4.0General biochemistryCrossref
Discovery of a molecular glue inhibitor that stabilises a non-productive Kalirin-Rac1 complex
J. L. Gray, M. C. Callens, A. Henley +16
bioRxiv8 Sep 2026PreprintAll rights reservedFull textMolecular biology & gene regulationbioRxiv
Bridging MS2 Spectra and Chemical Space: Advances in Spectral Similarity, Molecular Retrieval, and De Novo Structure Discovery
Julian Pollmann, Roman Bushuiev, Anton Bushuiev +2
ChemRxiv9 Sep 2026PreprintCC BYFull textMolecular biology & gene regulationChemRxiv
Leveraging high-throughput proteomics and AI-based protein folding to accelerate VAV1 molecular glue discovery.
Hanfeng Lin, Xin Yu, Haiyang Zheng +12
Nature communications12 Sep 2026Peer-reviewedCC BY-NC-NDFull text5 citedMolecular biology & gene regulationEurope PMC
Accelerating characterization of conformational stability for biologics in Discovery space by combining SEC‑HDX‑MS with molecular modeling
Zhuo Gao, Elizabeth F. Bayne, Michael Hartmann +3
ChemRxiv22 Oct 2026PreprintCC BYMolecular biology & gene regulationChemRxiv

Cite

BibTeX RIS

BibTeX

@misc{jakobsen2026informationdense,
  title = {{Information-Dense Synthesis for Molecular Discovery}},
  author = {Kasper K. Jakobsen and Eli N. Weinstein},
  howpublished = {arXiv},
  year = {2026},
  month = oct,
  doi = {10.48550/arxiv.2610.08495},
  eprint = {2610.08495},
  archivePrefix = {arXiv},
  url = {https://arxiv.org/abs/2610.08495},
  note = {CC BY 4.0},
}

RIS

TY  - UNPB
TI  - Information-Dense Synthesis for Molecular Discovery
AU  - Kasper K. Jakobsen
AU  - Eli N. Weinstein
PY  - 2026
DA  - 2026/10/06
PB  - arXiv
DO  - 10.48550/arxiv.2610.08495
UR  - https://arxiv.org/abs/2610.08495
AB  - Machine learning can accelerate molecular discovery by designing molecules and planning experiments. However, many scientific challenges demand molecules with very rare properties, and in this sparse setting, existing algorithms offer little gain over random guessing. We propose a method to efficiently search large regions of molecular space using algorithmically controlled stochastic synthesis. Rather than design, make and test individual molecules, we design and make complex mixtures, test them as a pool, then deconvolute the molecule-activity map. We optimize synthesis to encode maximal information. Theoretically, this approach can reduce the number of experiments required to find the optimal molecule among d candidates from 𝒪(d) to 𝒪(log d) or 𝒪(1). In simulation, on estimated protein fitness landscapes, it finds active molecules with an order of magnitude fewer experiments than existing Bayesian optimization methods.
ER  -

Abstract

Related

Cite