Matt, I'll check with the owner of the data about sending it to you. Could you let me know how's best to send it? Feel free to contact me off the list at this address.
The spectral comparison software sounds interesting! Thanks, DT Matthew Chambers wrote: > Can you discern the order the scans are enumerated in the PKLs with only > the precursor mass and the peaks? I.e. is it: > function=2 scan=1 > function=2 scan=2 > function=2 scan=3 > function=3 scan=1 > function=3 scan=2 > function=3 scan=3 > > or > > function=2 scan=1 > function=3 scan=1 > function=2 scan=2 > function=3 scan=2 > function=2 scan=3 > function=3 scan=3 > > I would expect one of those two orders (with the added complication of > scan combination: does it combine only within a single function or can > it also combine scans from multiple functions?). If it's some seemingly > random order, then it's indeed pretty hopeless to directly compare the > exports. :( > > However, even in the worst case it might be informative or at least > interesting to see what a spectral-duplicate-finder application could do > for comparing the two spectrum lists. Presumably, the combined scans in > the PKLs would score highest with the several uncombined scans in the > mzML that went into the combination. Deisotoping might throw a wrench in > the duplicate comparison, but the question would be does it throw a > smaller wrench into the comparison between the combined PKL scan and its > uncombined mzML counterparts than it does for other comparisons. We have > an as-yet unpublished tool that has several spectral duplicate scoring > algorithms, so I'll still take a look at the data if you want. We have > Waters raw data but none of it is processed with PLGS to compare > msconvert's processing to, so that's my motivation to help. :) > > -Matt > > > Dave Trudgian wrote: > >> Matt, >> >> Thanks for your comments. Unfortunately the pkl format doesn't retain >> any scan numbering or retention time information. I've tried comparing >> the output, but it's pretty much impossible to match things up. Trying >> to match precursor masses to compare associated MS/MS spectra doesn't >> work, due to the PLGS mass correction resulting in different precursor >> masses in the pkl vs mzXML, even if the RAW file is lockmass corrected >> with MassLynx prior to mzXML conversion. Thanks for the offer of having >> a look though. >> >> Nice to hear that PLGS should feature mzML export soon. We're eagerly >> awaiting PLGS 2.4 for many reasons. A new Aglient QTOF arrives here >> soon, and the Waters machine will then most likely be used mainly for >> label-free MSe work with PLGS, but would be nice to have the option of >> putting Waters DDA data through our pipeline. We now have 5 instruments >> from 5 different vendors... all good fun! >> >> Many Thanks, >> >> DT >> >> Matt Chambers wrote: >> >> >>> It usually comes down to spectrum processing - specifically scan >>> combination, deisotoping, and centroiding. MassWolf and msconvert are >>> unlikely to be able to do as well as the vendor's own processing without >>> significant effort devoted toward optimizing on that platform. I'll take >>> a look at the data if you want: I'd compare the results of the PKL >>> export (assuming they are named in a way that retains function/scan >>> information, or at least retention time) to msconvert's mzML export >>> (which I know preserves function/scan information) to try to determine >>> what is the difference in processing of the data. I know they're working >>> on an mzML export for PLGS; if they preserve scan combination metadata >>> in their output then that would be a great help as well. >>> >>> -Matt >>> >>> >>> dctrud wrote: >>> >>> >>> >>>> Dear all, another query r.e. Waters QTOF data.... >>>> >>>> We have a QTOF Premier which is now being used for SILAC experiments, >>>> and we want to quantify using ASAPRatio / Xpress. Have no problems >>>> doing this in our pipeline using Orbitrap data, but the QTOF data is >>>> causing headaches. Up until now we've mostly been using MassWolf just >>>> to get MS data into .mzXML for purposes other than ID & quant, PLGS >>>> having been the preferred analysis tool for the instrument. >>>> >>>> If we use .mzXML files and feed them to Mascot/OMSSA/Tandem + TPP we >>>> cannot get anywhere near the number of IDs as with .pkl files >>>> generated using PLGS. We've tried conversion to mzXML using MassWolf >>>> and msconvert, applying lockmass correction in MassLynx prior to >>>> conversion, and comparing acquisition of MS/MS scans in centroid mode, >>>> vs acquisition in profile mode converted to centroid using MassLynx. >>>> An example of the differences in IDs: >>>> >>>> RAW->PLGS->PKL->Mascot->TPP: 874 peptide IDs (1% FDR) >>>> RAW->MassWolf->mzXML->mgf->Mascot->TPP: 279 peptide IDs (1 % FDR) >>>> >>>> No matter what combination of converter / pre-processing in MassLynx >>>> we try, the number of IDs is never above 1/3 of that we get from a PKL >>>> file. We've also tried changes to the acquisition methods, without >>>> success. >>>> >>>> Is anyone able to provide a MassLynx method file, and/or procedure for >>>> pre-processing / .mzXML conversion they are using successfully with a >>>> QTOF Premier or similar instrument? >>>> >>>> Any help would be gratefully appreciated. Many Thanks, >>>> >>>> DT >>>> > > > > --~--~---------~--~----~------------~-------~--~----~ You received this message because you are subscribed to the Google Groups "spctools-discuss" group. To post to this group, send email to [email protected] To unsubscribe from this group, send email to [email protected] For more options, visit this group at http://groups.google.com/group/spctools-discuss?hl=en -~----------~----~----~----~------~----~------~--~---
