Metabolomics has emerged as an invaluable tool for prognostic and diagnostic purposes, last in the cascade of others OMICS -genomics, transcriptomics, and proteomics. Omics training usually covers experiment design, data generation, and collection, data preparation, data analysis, and the last but not the least - data interpretation.
At the end of this meticulous energy, time, and financial-consuming path, it might be totally none sense to fail to put your results into the broader biological context.
For those like me that have never been trained to interpret metabolomics data, how can we make sure to not miss important points? Is Basic knowledge in Biochemistry, Physiology, or physiopathology of the disease of your interest, enough to harness the full potential of metabolomics technologies for biomarker screening u.a?
I would like to discuss with experts out there, the most important assets for a right and successful data interpretation of metabolomics data.
Thank you for sharing your experience in the Metabolomics journey as well.
Only advice I can give is to have a background in chemistry so you can identify what chemical groups make sense and to be able to investigate related chemical families or to predict parent compounds. For example there may very little on a glycoside but if you look up the parent compound as an alcohol or methoxy ether you can often find a trove of relevant literature. Read all the papers you can about the annotation for the compound as there can often be non-enzymatic routes to the formation of compounds (especially when ROS is involved). Look up other synonyms for the compound as these are not standardized, especially in older literature. Use KEGG to help visualize the pathways where possible and always doubt the automatic annotations. I have seen nature papers where their "marker" for a cancer type is actually a tropical plant alkaloid only produced in one family of trees from southeast Asia (the patient samples from that study were not even from that region of the world). Also if you think the metabolite is important you must always do an alternative assay to verify that it is the correct annotation.
Thank you very much for your contribution. Using basic knowledge in chemistry to understand metabolites in a more global vision of their biological involvement is a very interesting approach. This directs me towards a more advanced interpretation of certain metabolites identified as differential in my case, but whose complex chemical composition was misleading. I tend to simply attribute their presence to solvent, food, or whatever, so thank you for this inspiring new approach.
I am afraid being 100% sure about the specificity of a compound's route formation is a dream we can barely afford in Metabolomics.
Read, Read, Read to be aware of all existing non-enzymatic routes and cross-check the automatic annotations...well noted. Thank You very much
If you were to list some of of the metabolites you are having problems interpreting, I could always offer an opinion as to what I think it means.