Hello everyone. Currently, I am working with a cancer xenograft model and I wanted to carry on the metabolomics and lipidomics study of tissues (e.g., tumor, liver and kidney) to evaluate the drug effect. However, to process the data, I need to normalize the data. How can I normalize the data for these tissues? Normally, in case of plasma and urine samples, we normalize using the the IS and creatinine value, respectively.
TIA.
Hi Md. Mamunur Rashid,
Typically fresh weight basis (mg tissue) is a good start, some people also do based on total protein (which is unreliable to me!: esp. when one does lipidomics and extracts in organic non-polar solvents- so unsure how proteins can be measured!); but then what has worked OK is normalizing based on internal standard used per lipid class or metabolite class, once data is acquired. Some people also try by using total RNA or DNA etc. which are also unreliable (can you trust the comaprability of RNA extraction with lipid extraction?). Also, why protein normalziation fails, because if gene expression to proteome correlation (in terms of abundances) across cell-types and tissues and organisms is roughly 0.25- 0.5 at max, how to correlate and why to do so, for proteins and metabolites?? Given there are 100s of PTMS and protein regulatory processes and metabolite accumulation/ depletion mechanisms on going etc.
Again, I repeat: but then what has worked OK is normalizing based on internal standard used per lipid class or metabolite class, once data is acquired. Also, having good and multiple QCs (that can be used for systemic corrections), good design of experiment, reducing batch effect etc., you can generate better data !
I am afraid, but the field is still grappling with such issues, and vendors trying to come up with normalization procedures for tissues etc. but there is no single or GOOD answer at all.
Thanks,
As mentioned by others weight basis is good start. However, if you use non-targeted metabolomics, normalize by total ion chromatogram intensity. Normalization is totally dependent on what type of samples you have and if you use internal standard or not during sample prep.
Dear Mamunur,
these references may be of some help:
Fattuoni C, et al. Preliminary metabolomics analysis of placenta in maternal obesity. Placenta, 2018;61:89-95. doi: 10.1016/j.placenta.2017.11.014
W.B. Dunn, et al., The metabolome of human placental tissue: investigation of first trimester tissue and changes related to preeclampsia in late pregnancy, Metabolomics 8 (2012) 579.
- Badges
- Science method