Feature request.

[andyzim111]

Thank the developer for providing so a fantastic tool for DIA data analysis.
I want to request two features of DIA-NN.

1. The number of peptides of identified protein in protein intensity matrix.
2. The modification site on protein in the precursor outputs(results.tsv and pr.matrix.tsv).

Thanks very much.

[vdemichev]

Hi, thank you for the suggestions!

1: Yes, can be done, but also easy to extract this info using R or Python. Something like:

df <- diann_load('report.tsv')
pep <- unique(df[df$Global.Q.Value <= 0.01,c('Stripped.Sequence','Protein.Group')])
counts <- table(pep$Protein.Group)

This is more flexible, as you can use your own way of filtering the data (e.g. can discard peptides with high CV values).

2: Seems often requested, the difficulty is when the peptide is mapped to multiple protein. In general, it's not difficult to just output all positions, I am pretty sure we will implement it at some point, but for now cannot give any timeframe.

Best,
Vadim

[vdemichev]

Visualisation of peptide positions is now possible with https://github.com/MannLabs/alphamap

[andyzim111]

Visualisation of peptide positions is now possible with https://github.com/MannLabs/alphamap

Thanks a lot.

[volvox89]

Hi thank you for your effort.
Are you by any chance planning on combining the isoform information into only 1 entry, or making the option to choose between isoform and canonical ID outputs?
Cheers

[vdemichev]

I am not sure, what exactly do you suggest as output?

[volvox89]

Hi, sorry for being not precise enough.
For the protein group output, DIA-NN often generates several outputs per Gene, persumably where a peptide can be mapped to several isoforms of the same protein (unless I misunderstood this part). I would love to be able to only get one protein ID per Gene, either by excluding these ambigous peptides, or by combining them into one quantification result.