{
  "abstract": "Cyclic peptides are promising modalities for targeting intracellular sites; however, cell-membrane permeability remains a key bottleneck, exacerbated by limited public data and the need for well-calibrated uncertainty. Instead of relying on data-eager complex deep learning architecture, we propose a monomer-aware decoupled global alignment kernel (MD-GAK), which couples chemically meaningful residue-residue similarity with sequence alignment while decoupling local matches from gap penalties. MD-GAK is a relatively simple kernel. To further demonstrate the robustness of our framework, we also introduce a variant, PMD-GAK, which incorporates a triangular positional prior. As we will show in the experimental section, PMD-GAK can offer additional advantages over MD-GAK, particularly in reducing calibration errors. Since our focus is on uncertainty estimation, we use Gaussian Processes as the predictive model, as both MD-GAK and PMD-GAK can be directly applied within this framework. We demonstrate the effectiveness of our methods through an extensive set of experiments, comparing our fully reproducible approach against state-of-the-art models, and show that it outperforms them across all metrics.Scientific contributionWe introduce monomer-aware decoupled global alignment kernels for Gaussian processes (MD-GAK and position-aware PMD-GAK) that align cyclic peptides at the sequence level using chemically rich monomer fingerprints and explicit positional priors, yielding positive-definite similarity measures tailored to permeability modeling. Compared with order-agnostic fingerprint methods, standard global-alignment kernels and state-of-the-art graph and language-model baselines, our alignment-aware GPs provide improved discrimination, probabilistic calibration and scaffold-level robustness under stringent, leakage-controlled cyclic-peptide permeability benchmarks.",
  "abstract_shown": true,
  "affiliations": [
    "Center for Applied Intelligent Systems Research in Health, Halmstad University, Kristian IV:s väg 3, 30118, Halmstad, Sweden. ali.amirahmadi@hh.se",
    "Molecular AI, Discovery Sciences, R&D, AstraZeneca, Pepparedsleden 1, 43183, Mölndal, Sweden. ali.amirahmadi@hh.se",
    "Division of Systems and Synthetic Biology, Department of Life Sciences, Chalmers University of Technology, Kemigården 1, 41296, Gothenburg, Sweden",
    "Molecular AI, Discovery Sciences, R&D, AstraZeneca, Pepparedsleden 1, 43183, Mölndal, Sweden",
    "Medicinal Chemistry, Research and Early Development, Respiratory & Immunology, BioPharmaceuticals R&D, AstraZeneca, Pepparedsleden 1, 43183, Mölndal, Sweden",
    "Center for Applied Intelligent Systems Research in Health, Halmstad University, Kristian IV:s väg 3, 30118, Halmstad, Sweden",
    "Department of Research and Development (FoU), Region Halland, Södra vägen 9, 30118, Halmstad, Sweden",
    "Centre for Environmental and Climate Science, Computational Science for Health and Environment, Lund University, Kontaktvägen 10, 22362, Lund, Sweden"
  ],
  "article_type": null,
  "arxiv": null,
  "author_total": 6,
  "authors": [
    {
      "affs": [
        0,
        1
      ],
      "name": "Ali Amirahmadi",
      "orcid": "0000-0002-1999-8435"
    },
    {
      "affs": [
        2,
        3
      ],
      "name": "Gökçe Geylan",
      "orcid": "0000-0001-5961-4953"
    },
    {
      "affs": [
        4
      ],
      "name": "Leonardo De Maria",
      "orcid": "0000-0002-8061-4242"
    },
    {
      "affs": [
        5,
        6
      ],
      "name": "Farzaneh Etminani"
    },
    {
      "affs": [
        5,
        7
      ],
      "name": "Mattias Ohlsson",
      "orcid": "0000-0003-1145-4297"
    },
    {
      "affs": [
        3
      ],
      "name": "Alessandro Tibo",
      "orcid": "0000-0002-9070-740X"
    }
  ],
  "bibtex": "https://research.draconitic.com/paper/96vx2ahnb6ry4.bib",
  "cited_by": 6,
  "doi": "10.1186/s13321-026-01276-5",
  "fulltext_url": null,
  "id": "96vx2ahnb6ry4",
  "issns": [
    "1758-2946"
  ],
  "landing_url": "https://doi.org/10.1186/s13321-026-01276-5",
  "language": "en",
  "licence": "CC-BY",
  "licence_label": "CC BY",
  "licence_source": "journal",
  "licence_url": "https://creativecommons.org/licenses/by/4.0/",
  "metadata_licence": null,
  "native_subjects": [
    [
      "openalex",
      "3002|Computational Drug Discovery Methods"
    ]
  ],
  "openalex": "W4416791640",
  "pdf_url": null,
  "pmcid": "PMC13439769",
  "pmid": "42557596",
  "published": "2026-08-04",
  "publisher": "BioMed Central",
  "ris": "https://research.draconitic.com/paper/96vx2ahnb6ry4.ris",
  "sources": [
    {
      "first_seen": "2026-11-14T13:20:23Z",
      "native_id": "W4416791640",
      "source": "openalex",
      "url": "https://doi.org/10.1186/s13321-026-01276-5"
    }
  ],
  "subjects": [
    {
      "field": "chemistry",
      "origin": "openalex:3002|Computational Drug Discovery Methods",
      "subfield": "medicinal"
    }
  ],
  "title": "A decoupled alignment kernel for peptide membrane permeability predictions",
  "title_text": "A decoupled alignment kernel for peptide membrane permeability predictions",
  "type": "peer-reviewed",
  "updated_at": "2026-11-16T13:20:50Z",
  "url": "https://research.draconitic.com/paper/96vx2ahnb6ry4",
  "venue": "Journal of Cheminformatics",
  "versions": []
}