Redundant column names in TMT combined_annotation.txt file

[ginnyintifa]

In the TMT workflow, in the combined_annotative.tsv.

Some columns are redundant.

1 Replicate. We are not allowing user to have duplicated sample IDs i.e. label column. so the replicate isn't helpful.every row would have value 1 in this column. I don't know if this column would be involved in any calculation in the functionalities.

2 experiment/plex I think we can just keep one of them, preferrably experiment.

[hsiaoyi0504]

For the second point, although it's redundant, keeping this gives users a little bit more flexibility for customizing the plot.

[hsiaoyi0504]

For the first point, I think they are for different purposes. [sample ID]_1 and [sample ID]_2 are technical replicates. The replicate column, however, is initially designed for bioreplicates as we have in the FragPipe manifest.

[hsiaoyi0504]

But it's actually confusing, for example, this is FragPipe manifest from ccRCC discovery DIA data: dia_manifest (1).fp-manifest.txt The experiment column is [Condition]_[sample num], and the replicate column here becomes more like technical replicates.

[hsiaoyi0504]

Updated in https://github.com/MonashProteomics/FragPipe-Analyst/commit/d17fbf9cdd388c8ae284599b64b19a3b5feb4913..

Now the combined_annotation.txt will looks like this	plex	channel	label	condition	replicate
3	126	CPT0148080004	Tumor	1
3	127N	CPT0148100003	Normal	1
3	127C	CPT0083100003	Tumor	1
3	128N	CPT0083140003	Normal	1
3	128C	CPT0118160003	Tumor	1
3	129N	CPT0118180003	Normal	1
3	129C	CPT0105510003	Tumor	1
3	130N	CPT0105550003	Normal	1
3	130C	TumorOnlyIR03
3	131N	pool03	pool	3
7	126	CPT0109680004	Tumor	1
7	127N	CPT0109700003	Normal	1
7	127C	CPT0105940003	Tumor	1
7	128N	CPT0105980003	Normal	1
7	128C	CPT0090170004	Tumor	1
7	129N	CPT0090200003	Normal	1
7	129C	CPT0105330003	Tumor	1
7	130N	CPT0105370004	Normal	1
7	130C	CPT0148080004	Tumor	2
7	131N	pool07	pool	7

[ginnyintifa]

Looking good! And do users need to change column names in the report now?

[hsiaoyi0504]

That's correct. Please let me know if you find any issue. Thanks for the contribution!

[hsiaoyi0504]

@ginnyintifa I would like re-open this. Actually, FragPipe doesn't allow users to have the same label for different samples anymore. Therefore, the user needs to indicate _1, _2 in the label column, and the shared prefix before _1, _2 is used to differentiate samples from different patients.

[ginnyintifa]

@ginnyintifa I would like re-open this. Actually, FragPipe doesn't allow users to have the same label for different samples anymore. Therefore, the user needs to indicate _1, _2 in the label column, and the shared prefix before _1, _2 is used to differentiate samples from different patients.

Really? in TMT workflow, duplicated labels in the same plex is not allowed. But if there are 2 name labels in different plexes, it is allowed right? But this would cause duplicates seen in TMT-I report.

[hsiaoyi0504]

Yes, but what I am talking about is which should be the standard way annotating samples. I still keep the logic for handling duplicate labels, but probably we should use _ as the standard way. Any thoughts?