Metabolomic Data Fusion for Colorectal Cancer Biomarkers

Colorectal-cancer research can draw on several kinds of information from blood. These data bring together protein biomarkers, fluorescence patterns, and nuclear magnetic resonance (NMR) metabolic profiles to support research into whether combining measurement types provides complementary information. The collection contains 94 processed plasma samples organized as 47 matched pairs. Each pair includes one person with verified colorectal cancer and one person with a colorectal adenoma, matched by age, gender, and the location of the tumour or adenoma. Adenomas are non-cancerous bowel growths that can sometimes develop into cancer. Datafusionnetdata.mat contains a 94 × 476 matrix divided into three blocks: two protein biomarkers: CEA and TIMP-1; 19 processed fluorescence features obtained through PARAFAC modelling; 455 integrated proton-NMR peak variables, comprising 201 CPMG and 254 NOESY-Presat variables. The biomarker values are log2-transformed and adjusted using matched no-finding controls, so they should not be interpreted as raw concentrations. Class information provides diagnosis, matched-pair identifier, age, and gender. Sample labels are coded. The plasma samples were collected under approval from Ethics Committee 01.080/03 and the Danish Data Protection Agency, reference 2003-41-3312. The collection is intended for research and method development, not as a clinically validated diagnostic test. The fluorescence block derives from citrate-plasma EEM measurements. Rayleigh and second-order scatter regions were removed, daily water-Raman measurements were used for intensity calibration, and diluted samples were background-corrected with phosphate-buffered saline. Original fluorescence dataset page: https://ucphchemometrics.com/2023/05/04/fluorescence-on-blood-plasma-for-cancer-diagnosis/ Original ERDA package: https://sid.erda.dk/share_redirect/bLmW34gkI2/Metabolomic%20data%20fusion.zip If you use these data, please cite:Bro, R.; Nielsen, H. J.; Savorani, F.; Kjeldahl, K.; Christensen, I. J.; Brünner, N.; Lawaetz, A. J. (2013). Data fusion in metabolomic cancer diagnostics. Metabolomics, 9(1), 3-8. https://doi.org/10.1007/s11306-012-0446-0Lawaetz, A. J.; Bro, R.; Kamstrup-Nielsen, M. H.; Christensen, I. J.; Jørgensen, L. N.; Nielsen, H. J. (2012). Fluorescence spectroscopy as a potential metabonomic tool for early detection of colorectal cancer. Metabolomics, 8(S1), 111-121. https://doi.org/10.1007/s11306-011-0310-7