Hi,
I just had a look at the results within a feature group. The retention time differences within the feature groups are similar (as expecxted). I found it strange however, to see that the fold change differences for features within a feature groups differed between the datasets analysed, whereas I had expected that features coming from the same analyte would give similar changes.
Koen
Yes, ideally the fold changes within one feature group should be very similar.
However, all these features get integrated individually and especially smaller features might be more noisy and therefore more challenging to integrate.
In a later processing step, features in the down-regulated group get filled in with fillPeaks().
All these steps happen before annotation.
The fold change itself is calculated as a mean fold change of the features from your two sample groups.
For a reduced version of the diffreport, I would suggest using the fold change of the feature with the highest intensity (probably [M+H]+) within that feature group.
Ralf.