Skip to content
DraconiticResearch
Search Sources API
203,203 studies31 sourcesUpdated 37 min ago
PreprintCC BYChemRxiv

Small-Molecule Data Splitting for Machine Learning in Drug Design

  • Cao-Minh Truong1,2
  • Van-Thinh To1,2
  • Gia-Bao Truong1,2
  • Alexey Orlov3,4
  • Pedro J. Ballester5,6
  • Olivier Taboureau1,2
  • Viet-Khoa Tran-Nguyen1,2
6 affiliations
  1. 1Université Paris Cité
  2. 2INSERM ERL
  3. 3CNRS UMR
  4. 4University of Strasbourg
  5. 5Department of Bioengineering
  6. 6Imperial College London
Open at sourcePDFDOICite

Abstract

Data splitting is a critical determinant of machine-learning performance in small-molecule drug design. Different strategies impose distinct assumptions about structural similarity, temporal order, data provenance and chemical-space coverage, thereby affecting both evaluation realism and the interpretation of model performance. This perspective critically examines major splitting approaches, discusses their advantages, applicability and limitations, and offers evidence-based considerations for designing partitions aligned with intended prediction scenarios.

Related

  1. Machine learning for RNA–small molecule interactions: methodologies, evaluation, and future directions
    Song Chen, Zhijian Huang, Yucheng Wang +5
    Briefings in BioinformaticsPeer-reviewedCC BY 4.0Full textMedicinal chemistry & drug discoveryPMC
  2. Docking, Molecular Dynamics, QSAR, and Machine Learning in Alzheimer's Small-Molecule Drug Discovery: Current Utility, Key Limitations, and Reproducibility Challenges
    Poulami Saha, Anuja Chouhan
    ChemRxivPreprintCC BYFull textGeneral chemistryChemRxiv
  3. Small Organic Molecule-Based Multicolor Fluorescent Smart Materials: From Design Principles and Optical Properties to Versatile Applications
    Zhen Yan, Taihong Liu, Liping Ding +1
    ACS Central SciencePeer-reviewedCC BY5 citedMaterials & polymersOpenAlex
  4. SMDD-Bench: Can LLMs Solve Real-World Small Molecule Drug Design Tasks?
    Kevin Han, Rui Zhang, Kathy Y. Wei +3
    arXiv (Cornell University)PreprintCC BYFull text6 citedMedicinal chemistry & drug discoveryOpenAlex
  5. Computational design and prioritization of database-absent small-molecule GLP-1 receptor agonist candidates from public bioactivity data
    JOGU MOHAN
    ChemRxivPreprintCC BYFull textGeneral chemistryChemRxiv

Cite

BibTeXRIS
BibTeX
@misc{truong2026smallmolecule,
  title = {{Small-Molecule Data Splitting for Machine Learning in Drug Design}},
  author = {Cao-Minh Truong and Van-Thinh To and Gia-Bao Truong and Alexey Orlov and Pedro J. Ballester and Olivier Taboureau and Viet-Khoa Tran-Nguyen},
  howpublished = {ChemRxiv},
  year = {2026},
  month = sep,
  doi = {10.26434/chemrxiv.15009691/v1},
  url = {https://chemrxiv.org/doi/full/10.26434/chemrxiv.15009691/v1},
  note = {CC BY},
}
RIS
TY  - UNPB
TI  - Small-Molecule Data Splitting for Machine Learning in Drug Design
AU  - Cao-Minh Truong
AU  - Van-Thinh To
AU  - Gia-Bao Truong
AU  - Alexey Orlov
AU  - Pedro J. Ballester
AU  - Olivier Taboureau
AU  - Viet-Khoa Tran-Nguyen
PY  - 2026
DA  - 2026/09/30
PB  - ChemRxiv
DO  - 10.26434/chemrxiv.15009691/v1
UR  - https://chemrxiv.org/doi/full/10.26434/chemrxiv.15009691/v1
AB  - Data splitting is a critical determinant of machine-learning performance in small-molecule drug design. Different strategies impose distinct assumptions about structural similarity, temporal order, data provenance and chemical-space coverage, thereby affecting both evaluation realism and the interpretation of model performance. This perspective critically examines major splitting approaches, discusses their advantages, applicability and limitations, and offers evidence-based considerations for designing partitions aligned with intended prediction scenarios.
ER  -