Machine Learning for High-Throughput Reaction Yield Prediction in DNA-Encoded Library Synthesis

Abstract

DNA-encoded library (DEL) synthesis relies on efficient DNA-compatible reactions to ensure library fidelity and the reliability of downstream affinity selection. However, most existing reaction yield prediction models were developed for general organic synthesis and typically use full-atom representations of complete reactants and products. In DEL single-cycle synthesis validation, an appropriate model reaction is to focus on the exposed reactive handle, its local chemical environment, the incoming building block (BB), and the reaction type, rather than large constant backgrounds such as the DNA tag, linker, and pre-existing scaffold. Here, we propose a functional-group-centric machine learning framework for DEL single-cycle reaction yield prediction. The model uses only the exposed substrate functional group (FG) and its coarse-grained local environment, the incoming BB, and the reaction type as inputs. Benchmarking on a real-world high-throughput DEL single-cycle validation dataset shows that graph neural network (GNN) models based on this localized representation outperform traditional fingerprint-based baselines, with a graph attention network (GAT) achieving the best overall performance and showing strong robustness on highly imbalanced industrial data. These results indicate that foregrounding the local reaction center is not merely a simplification of molecular representation, but a task-aligned strategy that better reflects the chemistry and decision logic of DEL single-cycle transformations. This lightweight framework is readily compatible with practical DEL workflows and provides valuable supports on reagent quality control (QC), building block selection, high-fidelity library design, and retrospective analysis of anomalous results.

IMPLICIT LEARNING: REACTION OUTCOME PREDICTION FROM POSITIVE-ONLY DATA
Nick Wierich, Niklas Dobberstein, Frank Glorius
ChemRxiv28 Oct 2026PreprintLicence not statedGeneral chemistryChemRxiv
A machine-learning approach to high-throughput spatially resolved ARPES measurements
Masaki Imamura, Daijiro Koga, Kazutoshi Takahashi
Scientific Reports23 Oct 2026Peer-reviewedCC BY 4.0Full textMaterials & polymersCrossref
pdb2reaction: End-to-End Reaction-Path Elucidation from PDB Structures Using Machine-Learning Interatomic Potentials
Takuto Ohmura, Hajime Sato, Tohru Terada
ChemRxiv6 Sep 2026PreprintLicence not statedGeneral chemistryChemRxiv
Bridging the Literature-to-Laboratory Gap with Language Models and Active Learning for Reaction Feasibility Prediction and Autonomous Exploration
Haowen Zhong, Yi Yang, Jiekai Jia +7
ChemRxiv3 Nov 2026PreprintLicence not statedGeneral chemistryChemRxiv
Construction of a Curated Literature Dataset and Machine Learning-Guided Exploration of Supported Metal Catalysts for Low- to Intermediate-Temperature Reverse Water–Gas Shift Reaction
Umi Shin, Shinya Mine, Aritomo Yamaguchi +1
ChemRxiv2 Nov 2026PreprintLicence not statedGeneral chemistryChemRxiv

Cite

BibTeX RIS

BibTeX

@misc{tang2026machine,
  title = {{Machine Learning for High-Throughput Reaction Yield Prediction in DNA-Encoded Library Synthesis}},
  author = {Tao Tang and Li Gao and Sen Gao and Hongyao Zhu and Guansai Liu and Jin Li and Xuemin Cheng},
  howpublished = {ChemRxiv},
  year = {2026},
  month = sep,
  doi = {10.26434/chemrxiv.15008275/v1},
  url = {https://chemrxiv.org/doi/full/10.26434/chemrxiv.15008275/v1},
  note = {CC BY},
}

RIS

TY  - UNPB
TI  - Machine Learning for High-Throughput Reaction Yield Prediction in DNA-Encoded Library Synthesis
AU  - Tao Tang
AU  - Li Gao
AU  - Sen Gao
AU  - Hongyao Zhu
AU  - Guansai Liu
AU  - Jin Li
AU  - Xuemin Cheng
PY  - 2026
DA  - 2026/09/03
PB  - ChemRxiv
DO  - 10.26434/chemrxiv.15008275/v1
UR  - https://chemrxiv.org/doi/full/10.26434/chemrxiv.15008275/v1
AB  - DNA-encoded library (DEL) synthesis relies on efficient DNA-compatible reactions to ensure library fidelity and the reliability of downstream affinity selection. However, most existing reaction yield prediction models were developed for general organic synthesis and typically use full-atom representations of complete reactants and products. In DEL single-cycle synthesis validation, an appropriate model reaction is to focus on the exposed reactive handle, its local chemical environment, the incoming building block (BB), and the reaction type, rather than large constant backgrounds such as the DNA tag, linker, and pre-existing scaffold. Here, we propose a functional-group-centric machine learning framework for DEL single-cycle reaction yield prediction. The model uses only the exposed substrate functional group (FG) and its coarse-grained local environment, the incoming BB, and the reaction type as inputs. Benchmarking on a real-world high-throughput DEL single-cycle validation dataset shows that graph neural network (GNN) models based on this localized representation outperform traditional fingerprint-based baselines, with a graph attention network (GAT) achieving the best overall performance and showing strong robustness on highly imbalanced industrial data. These results indicate that foregrounding the local reaction center is not merely a simplification of molecular representation, but a task-aligned strategy that better reflects the chemistry and decision logic of DEL single-cycle transformations. This lightweight framework is readily compatible with practical DEL workflows and provides valuable supports on reagent quality control (QC), building block selection, high-fidelity library design, and retrospective analysis of anomalous results.
ER  -

Abstract

Related

Cite