Assessment of different machine learning tools employed in lipidomics

2nd International Conference on Chemo and Bioinformatics ICCBIKG 2023 (330-333)

АУТОР(И) / AUTHOR(S): David Pirić, Romana Masnikosa


Download Full Pdf   

DOI: 10.46793/ICCBI23.330P


Herein we present the potential of four machine learning (ML) algorithms: Partial Least Squares – Discriminant Analysis (PLS-DA), Random Forests (RF), Support Vector Machines (SVM) and Decision Trees (DT) to classify human plasma samples into cancer or control group, using a lipidomic dataset retrieved from the research article „Lipidomic profiling of human serum enables detection of pancreatic cancer“ by Wolrab et al. [1]. Our main objective was to assess and compare, for the four ML techniques, the performance metrics, that is accuracy, precision, sensitivity, F1 score and ROC-AUC, with those computed by Orthogonal Projections to Latent Structures (OPLS)-DA in [1]. Our findings suggest that SVM and RF offer a superior performance as binary classifiers, making these two promising candidates for future use in the discovery of potential lipidomic biomarkers.


machine learning, lipidome, classification, random forests, support vector machines


  • D. Wolrab et al., Lipidomic profiling of human serum enables detection of pancreatic cancer, Nature communications, 13 (2022), 124, doi: 10.1038/s41467-021-27765-9.