How does MetaboAnalyst calculate the p values associated with fold change analysis?

MetaboAnalyst
#1

Hi, I just need to ask about the p values associated with fold change analysis, what test does MetaboAnalyst use to calculate them? My data (from lipidomics analysis, about 2500 features) is not normally distributed, and after normalization with autoscaling, there are no significant features found btn groups using t-test or non-parametric testing. I did fold change analysis (unpaired) and many features were significant at a threshold of 2. So, I need to know what kind of statistical test do you rely on to do this?

Can you help me with this please?

#2

I am not aware of p values associated with FC analysis.

 I did fold change analysis (unpaired) and many features were significant at a threshold of 2.

It just means these features pass the threshold you specified. Nothing more.

#3

Thank you for the clarification, I got confused because they are called “sig. down” and “sig.up” features.

It makes sense now. Thanks a lot!