Normative Modeling of Plasma Metabolomics for Lung and Breast Cancer Detection in Never-Smokers: analysis notebook
Jupyter notebook implementing the analyses of "Normative Modeling of Plasma Metabolomics for Lung and Breast Cancer Detection in Never-Smokers" (Metabolites, 2026). Each subject's plasma metabolite profile is compared with cancer-free controls of the same cohort and smoking status matched on age and body mass index, and classifiers trained on these deviations and on the original metabolite levels are combined and compared with an XGBoost baseline under nested cross-validation. The notebook also contains the ablation, calibration, decision-curve, bootstrap, permutation, pathway and sensitivity analyses reported in the article. The participant-level data are not included; they are covered by an agreement with BioMark Diagnostics and are available from the corresponding author on reasonable request. The notebook is distributed with outputs cleared and runs on any dataset in the documented input format.
Authors
- Paramjit S. Tappia (ORCID: https://orcid.org/0000-0001-8307-2760)
- Maria L. Vaida (ORCID: https://orcid.org/0000-0002-7869-1900)
- Guoyu Huang (ORCID: https://orcid.org/0009-0004-5854-9482)
- Rashid Ahmed Bux
- Bram Ramjiawan (ORCID: https://orcid.org/0009-0004-3450-0887)
- Bharadwaj Popuri (ORCID: https://orcid.org/0000-0002-2979-5199)
- Jean-François Haince
- John Ray Martinez
Institutions
- Harrisburg University of Science and Technology (US)
- St. Boniface Hospital (CA)
- Biomark (United States) (US)
- University of Manitoba (CA)
Publication Details
- Journal
- Zenodo (CERN European Organization for Nuclear Research)
- Published
- 2026-11-21
- DOI
- https://doi.org/10.5281/zenodo.24429607
- Primary Topic
- Metabolomics and Mass Spectrometry Studies
- Type
- article
- Field-Weighted Citation Impact
- 0.00