wkumler
commented
20 hours ago Outliers do not seem to be due to the order of the measurements, though there was clearly drift in response factor over the course of the run that was controlled for by the internal standards:
Figure 2 lower panels colored by timestamp instead of by SLA:
Evidence of response factor drift over the course of the run in the internal standards:
Average metabolite peak area without BMIS normalization:
Average metabolite peak area after BMIS normalization:
Resolved with 310e6aa
Reviewer 2 | 04 Oct 2024 | 12:20 The high-quality analytics is appreciated. It should be verified if the variations between biological replicates could be caused by the order of measurements in the sequence. If >100 samples of 15 min are run in a row there could be a drift that would be hidden because of the randomization. It would be easy to have a look at the outliers (e.g. Fig. 2 lower panels) are measured early / late during the sequence.