{
  "abstract": "Cytochrome P450 (CYP) enzymes mediate the oxidative metabolism of most marketed drugs, making early prediction of their sites of metabolism (SoMs) essential for addressing metabolic liabilities. In this work, we present a substantial update to our hybrid docking/quantum mechanics SoM prediction tool IMPACTS (J. Chem. Inf. Model., 2012, 52, 2471-2483). This update is evaluated on newly curated, fully annotated datasets containing over 2,400 datapoints and spanning nine human isoforms; these datasets are released to the community in MOL2 and JSON formats. Technical implementations presented in this work include 1) development of a three-dimensional solvent-accessible surface area correction to the precomputed activation energies, 2) design and integration of a highly efficient L-BFGS optimizer, and 3) elimination of a filter that had been historically used to remove unproductive binding modes. We follow these updates with extensive benchmarking and analysis on the newly curated datasets, where we show 1) achievement of a global Top-2 accuracy of ~75% across the nine isoforms and 2) the effective computational cost using the new optimizer was less than 1 minute per metabolite for eight of the nine isoforms. Finally, we present six real-world case studies comprised of approved drugs and discontinued clinical candidates. These case studies illustrate both the practical value and the remaining limitations of our approach.",
  "abstract_shown": true,
  "affiliations": [
    "Molecular Forecaster Inc.",
    "Department of Chemistry, McGill University, 801 Sherbrooke Street West, Montreal, QC, Canada, H3A 0B8",
    "AbbVie, 381 Plantation Street, Worcester, MA, United States of America"
  ],
  "article_type": null,
  "arxiv": null,
  "author_total": 7,
  "authors": [
    {
      "affs": [
        0,
        1
      ],
      "name": "Mihai Burai-Patrascu",
      "orcid": "0000-0001-9289-7887"
    },
    {
      "affs": [
        0
      ],
      "name": "Antoine Moitessier"
    },
    {
      "affs": [
        0
      ],
      "name": "Matej Janezic"
    },
    {
      "affs": [
        0
      ],
      "name": "Anne Labarre"
    },
    {
      "affs": [
        2
      ],
      "name": "Kevin Cusack"
    },
    {
      "affs": [
        0
      ],
      "name": "Joshua Pottel"
    },
    {
      "affs": [
        0,
        1
      ],
      "name": "Nicolas Moitessier"
    }
  ],
  "bibtex": "https://research.draconitic.com/paper/fesh1qsybhj6s.bib",
  "cited_by": 0,
  "doi": "10.26434/chemrxiv.15009130/v1",
  "fulltext_url": null,
  "id": "fesh1qsybhj6s",
  "issns": [],
  "landing_url": "https://chemrxiv.org/doi/full/10.26434/chemrxiv.15009130/v1",
  "language": null,
  "licence": "CC-BY-NC-ND",
  "licence_label": "CC BY-NC-ND",
  "licence_source": "journal",
  "licence_url": "https://creativecommons.org/licenses/by-nc-nd/4.0/",
  "metadata_licence": null,
  "native_subjects": [
    [
      "openalex",
      "1703|Computational Drug Discovery Methods"
    ]
  ],
  "openalex": "W7213902158",
  "pdf_url": "https://chemrxiv.org/doi/pdf/10.26434/chemrxiv.15009130/v1",
  "pmcid": null,
  "pmid": null,
  "published": "2026-09-21",
  "publisher": "American Chemical Society (ACS)",
  "ris": "https://research.draconitic.com/paper/fesh1qsybhj6s.ris",
  "sources": [
    {
      "first_seen": "2026-10-30T12:17:19Z",
      "native_id": "10.26434/chemrxiv.15009130",
      "source": "chemrxiv",
      "url": "https://chemrxiv.org/doi/full/10.26434/chemrxiv.15009130/v2"
    }
  ],
  "subjects": [
    {
      "field": "chemistry",
      "origin": "source:chemrxiv",
      "subfield": "general"
    }
  ],
  "title": "Experimental Data Analysis and Improved Cytochrome P450 Site-of-Metabolism Prediction for Drug Discovery",
  "title_text": "Experimental Data Analysis and Improved Cytochrome P450 Site-of-Metabolism Prediction for Drug Discovery",
  "type": "preprint",
  "updated_at": "2026-10-30T12:17:21Z",
  "url": "https://research.draconitic.com/paper/fesh1qsybhj6s",
  "venue": "ChemRxiv",
  "versions": []
}