Skip to content
DraconiticResearch
Search Sources API
203,234 studies31 sourcesUpdated 10 min ago
PreprintCC BYChemRxiv

Automatic Text-fishing, Unification, and Normalization Application in Python (aTUNApy): a tool to build molecular databases

  • Aylin Del Moral Morales1,2,3,4
  • José L. Medina-Franco3,4
4 affiliations
  1. 1Ciencias Naturales
  2. 2Universidad Autónoma Metropolitana-Cuajimalpa (UAM-C)
  3. 31 Grupo de Investigación DIFACQUIM, Departamento de Farmacia, Facultad de Química
  4. 4Universidad Nacional Autónoma de México
Open at sourcePDFDOICite

Abstract

The construction and maintenance of chemical databases from the scientific literature remain labor-intensive, largely because relevant information is dispersed across thousands of publications and compounds are reported under inconsistent nomenclature. Here we present aTUNApy (automatic Text-fishing, Unification, and Normalization Application in Python), an open-source tool that couples large language model (LLM)-based extraction of compound information from PubMed abstracts with nomenclature harmonization through PubChem and SMILES standardization with RDKit. aTUNApy retrieves articles for a user-defined query, extracts compound names together with user-specified attributes such as bioactivity, mechanism or source, and maps each mention to a unique PubChem identifier, collapsing synonyms onto a single canonical entry. Four LLM families are supported: GPT, Claude, LLaMA and Gemini. We benchmarked all four against a manually curated reference set of 507 articles, including 100 negative controls, covering 148 unique compounds. Claude achieved the highest performance (precision 0.98, recall 0.95, F1 0.97), while Gemini reached comparable precision (0.98) and the lowest hallucination rate (0.2%) at no API cost, reducing a five-day manual curation task to minutes or hours of automated processing. Finally, to demonstrate end-to-end utility, we applied aTUNApy to 8,273 articles from PubMed and Web of Science to build NutriEpiDB, a database of 653 dietary compounds with reported epigenetic effects across 917 food sources.

Related

  1. Automatic Generation of Novel Highly-Functional Molecular Sequences
    Vincent R. Ragusa, Chrisoph Adami
    arXivPreprintCC BY-NC-ND 4.0Full textBiophysicsarXiv
  2. Assessing the impact of parental linear gene normalization on the performance of statistical models for circular RNA differential expression analysis
    Briefings in bioinformaticsPeer-reviewedCC BY-NCFull textMolecular biology & gene regulationEurope PMC
  3. Optimized Super-SILAC Normalization Enables Enhanced Profiling of Histone Post-Translational Modification Changes Associated with Microbiota and Acute Inflammation
    Dieggo Rodrigues de Paula, Alessandro Vai, Roberta Noberini +5
    Molecular & Cellular ProteomicsPeer-reviewedLicence not statedProteomics & metabolomicsCrossref
  4. Automated reconstruction of reproducible protein databases with RepDBmaker.
    Giacomo Mutti, Toni Gabaldón
    Protein science : a publication of the Protein SocietyPeer-reviewedCC BY-NCFull textMolecular biology & gene regulationEurope PMC
  5. Research progress and trends of gut microbiota and lung injury: a bibliometric and visualization analysis based on two databases (2015–2026)
    Xinru Wu, Tiancheng Yu, Shuai Hu +1
    Frontiers in medicinePeer-reviewedCC BYFull textMolecular biology & gene regulationEurope PMCEurope PMCPMC

Cite

BibTeXRIS
BibTeX
@misc{morales2026automatic,
  title = {{Automatic Text-fishing, Unification, and Normalization Application in Python (aTUNApy): a tool to build molecular databases}},
  author = {Aylin Del Moral Morales and José L. Medina-Franco},
  howpublished = {ChemRxiv},
  year = {2026},
  month = sep,
  doi = {10.26434/chemrxiv.15008687/v1},
  url = {https://chemrxiv.org/doi/full/10.26434/chemrxiv.15008687/v1},
  note = {CC BY},
}
RIS
TY  - UNPB
TI  - Automatic Text-fishing, Unification, and Normalization Application in Python (aTUNApy): a tool to build molecular databases
AU  - Aylin Del Moral Morales
AU  - José L. Medina-Franco
PY  - 2026
DA  - 2026/09/11
PB  - ChemRxiv
DO  - 10.26434/chemrxiv.15008687/v1
UR  - https://chemrxiv.org/doi/full/10.26434/chemrxiv.15008687/v1
AB  - The construction and maintenance of chemical databases from the scientific literature remain labor-intensive, largely because relevant information is dispersed across thousands of publications and compounds are reported under inconsistent nomenclature. Here we present aTUNApy (automatic Text-fishing, Unification, and Normalization Application in Python), an open-source tool that couples large language model (LLM)-based extraction of compound information from PubMed abstracts with nomenclature harmonization through PubChem and SMILES standardization with RDKit. aTUNApy retrieves articles for a user-defined query, extracts compound names together with user-specified attributes such as bioactivity, mechanism or source, and maps each mention to a unique PubChem identifier, collapsing synonyms onto a single canonical entry. Four LLM families are supported: GPT, Claude, LLaMA and Gemini. We benchmarked all four against a manually curated reference set of 507 articles, including 100 negative controls, covering 148 unique compounds. Claude achieved the highest performance (precision 0.98, recall 0.95, F1 0.97), while Gemini reached comparable precision (0.98) and the lowest hallucination rate (0.2%) at no API cost, reducing a five-day manual curation task to minutes or hours of automated processing. Finally, to demonstrate end-to-end utility, we applied aTUNApy to 8,273 articles from PubMed and Web of Science to build NutriEpiDB, a database of 653 dietary compounds with reported epigenetic effects across 917 food sources.
ER  -