Metabolomics BMD and mPOD using wild animals

[DLettoof]

I have a wild animal untargeted metabolomics dataset for 4 tissue types for assessing the metabolome against individual contaminant concentrations. I am hoping to estimate a metabolic point of departure based on the BMDs. During the feature selection step my first dataset find no features at FDR < 0.5, 7 between 0.51 - 0.63 and 2000 + at 0.64... I know this is a huge value, and with wild animals there is a lot of "noise" in the data and I believe the contaminant concentrations they're exposed to a relatively low.

I'm wondering if there is any reason to pursue further analyses based on a high FDR or it's not worth any biological interpretation...

I have done these analyses in MetaboAnalyst binning concentrations in "doses" with replicates of 3 min (not ideal) and selecting features based on raw p not FDR, but that isn't an option in Dromics.

Interested to hear some inputs?