search

rpeckner-broad / Specter

Specter: linear deconvolution for targeted analysis of data-independent acquisition mass spectrometry proteomics
17 stars 12 forks source link

Overlapping isolation scheme data #10

Open kaveelim opened 6 years ago

kaveelim commented 6 years ago

Hi,

I tried to use Specter to analyse my data with overlapping isolation scheme (https://skyline.ms/files/home/software/Skyline/2013-ASMS-Overlapped-DIA.pdf). The problem I found is the quantification turns out to be all zero. I think this is cause by failure to meet the requirement of 5 consecutive scans with non-zero coefficient. This occurred because after demultiplexing data (msconvert), one spectrum (e.g., 20-m/z wide windows) was split the recorded precursor m/z is split into 2 consecutive scans with the same precursor m/z. Is it possible to use 'isolation window' information instead of precursor m/z?

Thank you, Kavee

rpeckner-broad commented 6 years ago

Hi Kavee, Unfortunately, you're correct that the splitting of spectra into pairs with identical m/z values makes the quantification script fail. I can fix this shortly. In the meantime, there should be no trouble running the analysis with the non-demultiplexed versions of your mzMLs.