Isotope tracing, metabolomics, and lipidomics often default to generic statistical pipelines borrowed from transcriptomics, including volcano plots and GSEA-like enrichment against pathway databases. These tools are convenient, but they were built for gene lists, not stoichiometrically linked metabolic networks, and can obscure the biology they are meant to reveal. We argue that a biology-centric approach, such as ratio-based metrics, rather than those generated by an off-the-shelf pipeline, provides more meaningful outcomes that standard approaches miss, across three settings.
Firstly, we used pathway metabolite ratios as a biology-led analytical approach to identify distinctive glutamine handling across cancer cell lines that bulk differential and enrichment analyses failed to distinguish.
Next, we applied ¹³C-fatty acid tracing to show that fatty acids contribute markedly less carbon to the TCA cycle compared to glucose and glutamine in cancer cell models, and that extracellular fatty acid delivery via CPT1a is substantially redirected toward cardiolipin turnover rather than combustion in prostate cancer cells. A routing decision invisible to standard metabolomic approaches.
Finally, we extended this concept to whole-tissue lipidomics in a triple-negative breast cancer model on defined macronutrient diets, and determined that relative fatty acid saturation and chain-length ratios within specific phospholipid classes, rather than class-level or lipid species abundance, correlate with tumour mass and lung metastatic burden.
Across all three, we promote the benefits of biologically-centric analyses to large-scale data approaches.