Number of Proteins.Identified Vs proteins in report.pg_matrix

[witszymanski]

Hi all,
I have a small question because I can't really find the answer on the github page. In the report.stats.tsv I can find the numbers of identified proteins and all my samples have around 2100.

However, when I look at the report.pg_matrix.tsv table, I can find around 4000 rows and the distribution among samples looks like this:

(sample names are different because I renamed them in R)
Can you explain me these numbers?

Side question: is this the number of commonly used "points per peak" ?

All the best
Witek

[vdemichev]

Hi Witek,

• Filtering is different, also in the .stats file it's uniquely identified proteins that are counted.
• No, it's peak width as measured in SWATH cycles. The average number of data points per peak is printed by DIA-NN in the log when it's run with scan window = 0.

Best,
Vadim

[vdemichev]

I updated the docs with the description of how the number of proteins in the stats file is calculated

[witszymanski]

Filtering is different, also in the .stats file it's uniquely identified proteins that are counted.

Ok, Thanks! I'll check the new description!

No, it's peak width as measured in SWATH cycles. The average number of data points per peak is printed by DIA-NN in the log when it's run with scan window = 0.

Ok, I have the log files and it was ran with scan window = 0 . Where do I find it now? How is the line called?

[vdemichev]

'Peak width'