Open jpfeuffer opened 3 years ago
@jpfeuffer I read the document in the morning and is quite difficult to follow. I think is better in this thread here to discuss the metrics with the following structure:
In this way we can discuss easily the metrics to be added and is more clear for the students.
I agree. Will do.
As discussed I condensed the Table a little bit with only the most important information: https://docs.google.com/spreadsheets/d/1eHM6cCT0TYpG40pJ8wS3lFq6OHQwjPUEIzevThgfAQI
I still need to add info on where to get the information. The table is open for comments with everyone.
A large benchmarking study is trying out proteomicslfq together with pmultiqc. They are looking for a list of metrics. Maybe some of them could be directly added to pmultiqc.
https://docs.google.com/spreadsheets/d/1azv3MlW5PwDBBtIqXoTFrg00nI-5TIVrx1gKLMl81G0/edit?usp=sharing
What do you think? I am happy to help with where to get the data from and already made some comments in the Document. I think some of the metrics are already in there by now.
I am not sure about the ground-truth metrics, since they are of course specific to certain datasets but in theory, we could create a GitHub repo in a standardized way with expected fold changes or concentrations with reference to the sample names in the SDRF. Then pmultiqc could check if a ground-truth annotation is available for this dataset and add additional metrics.