Stable-isotope-resolved metabolomics (SIRM) feeds a 13C/15N/2H-labeled tracer and measures how label propagates into downstream metabolites, reporting metabolic ACTIVITY (flux) rather than pool size. This skill guards against the two errors that silently fabricate flux: reading a pool concentration as activity (pool and labeling can move in opposite directions), and interpreting raw isotopologue intensities without the mandatory natural-abundance / tracer-purity correction and steady-state check.
pip install isocor numpy
# high-resolution R alternative:
# R: install.packages('accucor')
Conceptual prerequisites: the chosen tracer and its purity, the molecular (and any derivatization) formula per metabolite, the instrument resolution (low-res QqQ vs high-res Orbitrap), and whether the design is single-timepoint (needs verified steady state) or a labeling time course.
Tell your AI agent what you want to do: - "Correct these raw isotopologue areas for natural abundance and tracer purity." - "Compute the MID and fractional enrichment for this metabolite." - "Help me choose a tracer for measuring glutamine carbon flux." - "Decide whether my single-timepoint labeling can be read as flux." - "Explain why my unlabeled control still shows an M+1 peak."
"I have M+0..M+5 areas for lactate from a U-13C6-glucose experiment on a triple quad. Apply natural-abundance and 99% tracer-purity correction and give me the corrected MID." "My GC-MS data is TBDMS-derivatized; correct the isotopologues accounting for the derivative formula."
"I want to know whether the pentose phosphate pathway or glycolysis dominates glucose handling in these cells - which tracer should I use and what readout?" "My intermediate pool went up but I think flux went down - how do I tell these apart with a tracer?" "I sampled labeling at 5, 15, and 30 minutes - is this at isotopic steady state, and can I run classical 13C-MFA?"