Could Libra give me a protein quantification matrix?

[JWang Lu]

Hello,

I am trying to analyse my TMT-labelled MS data using Libra. I have already got the pep.xml file from comet search and analysed all peptides by Libra. Libra has given me two files, one is a new pep.xml which contains several values for each peptide, and the other is a prot.xml file, which was output by ProteinProphet.

I hope to generate a protein quantification matrix, which contains absolute quantitative value for each protein and each TMT-label, convenient for my subsequent analysis and visualization in R or Python. 

But I have some problems here:

1) I hoped to use absolute values, for some peptide quantification in my reference group is zero, wich may affect my experiment group data. I used Normalize against sum of reagent profiles in Libra condition file, is that mean I could got the absolute values? Besides in the output tsv file from prot.xml, I got the ratio instead of a quantitative value, how could I got that number?

2) Another question is that maybe two or more protein share the same quantification result. I understand MS could not distinguish these protein with high sequence similarity, but could I output a report for every protein directly by TPP tools?

3) In my Libra output pep.xml, there're nearly a half peptide abandoned in protein quantification (kept column was set to No or blank). Is this normal or bad TMT-label? Furthermore, there're many proteins even without the ratio value.

How could I achieve my goal? If Libra could not output directly, which tools could I use or how could I calculate by myself? Thanks a lot!

Respectfully,

Jinwang Lu