Issues with non-continuous scan indices

[rickhelmus]

Hello,

I am facing some issues reading Agilent Q-TOF data that apparently has scan numbers that are not continuous (but sorted).

When readMZPFile() is called with scNum==0 then things are fine, mostly owing to the following line:

mstoolkit/src/MSToolkit/MSReader.cpp

Line 1380 in d652109

if (pScanIndex[rampIndex]<0) continue;

However, in my case I want to iterate over the file in chunks over several parallel workers and therefore need to set scNum at least once per chunk. However, there doesn't seem to be a way to actually find which scan numbers between 1 and getLastScan() are valid. I guess a way could be to export pScanIndex so the validity could be checked before attempting a read. But perhaps there are better ways to take of this, so I wanted to open this issue first before making any PR.

Thanks,
Rick

[mhoopmann]

Hi Rick,
Yes, index-based data files are difficult to work with. Using the scan index instead of the scan numbers is the right way to go, as you noticed. You will be able to get by from making a single pass on pScanIndex and finding the boundaries you want to use. But more to the point, it would be easier if you had a dense array of scan indexes instead of a sparse one for non-contiguous scan indexes. I'll leave the issue open, then circle back after ASMS and find a better solution.
Cheers,
Mike

[rickhelmus]

Great, that would be perfect!

Thanks,
Rick