Hi Eric, I'm sorry, if it's a little bit confusing. Theoretically I can rebuild the spectral libraries but it will be difficult to get all pepxml files. So I just want to know if it is risky to take this mixed approach.
It seems to be a thing of the mix up of older and newer spectrast version, because if I built a spectral library myself fully with spectraST 5.0 there are no m ion annotations no matter which import parameters I use. So, even if these m ions are more plausible they don't occur if I built it from the scratch. Also it's interesiting that these changes only occur during import from splib files, not from msp files. Further, libraries with m ion annotations change "back" to the m ion free annotation if they are imported from msp files. *(1) original:* Name: AAAALILAGLVADGK/2 301.1492 347.0 y6^2/-0.530,y3-17/-0.985,y3-18/-0.001 48/44 0.0057|0.58 311.1680 414.6 ? 50/44 0.0054|0.72 319.1597 1481.3 y3/-0.002 71/44 0.0025|0.32 327.2017 2003.3 ? 71/44 0.0048|0.30 341.2172 726.1 ? 59/44 0.0060|0.42 *(2) after import (1) from splib file with -cM parameter:* Name: AAAALILAGLVADGK/2 301.1506 347.0 y3-18/0.000 48/44 0.0057|0.58 311.1680 414.6 ? 50/44 0.0054|0.72 319.1612 1481.3 y3/0.000 71/44 0.0025|0.32 327.2027 2003.3 m2:5/0.000 71/44 0.0048|0.30 341.2183 726.1 m9:12/0.000 59/44 0.0060|0.42 *(3) after reimport (2) only from msp file (w or w/o -cM parameter):* Name: AAAALILAGLVADGK/2 301.6790 347.0 y6^2/0.000,y3-17/-0.456,y3-18/0.528 48/44 0.0057|0.58 311.1680 414.6 ? 50/44 0.0054|0.72 319.1612 1481.3 y3/0.000 71/44 0.0025|0.32 327.2027 2003.3 ? 71/44 0.0048|0.30 341.2183 726.1 ? 59/44 0.0060|0.42 All steps were done with the same version of spectraST. Especially the first line: where the software gets the information about the other alternative fragment types (the msp file was the only file in directory)? I'm sorry if this is confusing but I just want to understand what is happening during the processes. Thank you very much for your patience! Kind regards, Kristina Am Montag, 23. September 2019 17:48:10 UTC+2 schrieb Eric Deutsch: > > Hi Kristina, I am not so sure about answers to your questions. It would be > ideal to use SpectraST 5.0 in all your processing if you can easily do > this, yes, but in some cases it would not matter much. In general, where > peak interpretation is involved, SpectraST 5.0 is better. But I don’t know > relative to what you’re doing exactly when peak re-interpretation happens. > > > > Can you just rebuild all your libraries with SpectraST 5.0? That would be > safest. > > > > Regards, > > Eric > > > > > > > > *From:* [email protected] <javascript:> < > [email protected] <javascript:>> *On Behalf Of *Kristina Plate > *Sent:* Monday, September 23, 2019 4:22 AM > *To:* spctools-discuss <[email protected] <javascript:>> > *Subject:* Re: [spctools-discuss] Combination of consensus libraries and > import in MRM format via spectraST > > > > Hi Eric, > > > > thank you very much for your reply and the detailed and, yes, quite > reasonable explanation. Some of the spectral libraries I got for the > combination were built with SpectraST 4.0 and the remaining builds as well > as the combination and further import steps were done with SpectraST 5.0. > But there are two factors bothering me: > > > > 1. I thought too, that the "issue" is caused by different SpectraST > versions. But when I did the MRM import (spectraST 5.0) on the single > spectral libraries which were built with the older version (see above), > this phenomenon didn't occur. > > > > 2. Even if it's explainable by a smarter interpretation by SpectraST, does > that mean that I can use these changed spectra or is it questionable to use > spectral libraries of older software versions imported with a newer one? > > > > Thank you and kind regards, > > Kristina > > Am Freitag, 20. September 2019 20:49:28 UTC+2 schrieb Eric Deutsch: > > Hi Kristina, I’m not certain I fully understand your issue, but maybe I > can answer some questions: > > > > Partial old: > > Name: DIETIIQK/2 *(old entry, original library)* > > 308.1581 226.5 ? > > 309.0826 359.4 ? > > 325.6734 1596.8 b6-35^2/0.000,b6-36^2/0.492 > > 329.1945 141.2 a6^2/0.000 > > 343.1920 668.4 > b6^2/0.000,y3-45/-0.042,y6-45^2/-0.516,y6-46^2/-0.024 > > > > Partial new: > > Name: DIETIIQK/2 *(new entry, combined library)* > > 308.1581 226.5 ? > > 309.0826 359.4 ? > > 326.1680 1596.8 ? > > 328.2231 141.2 m4:6/0.000 > > 344.1816 668.4 m3:5/0.000 > > > > The first two peaks are fine. For the third entry, I am guessing that in > the original library this peak got shifted to exact m/z of the presumed > interpretation b6-35^2. But this is a highly unlikely interpretation. So in > the new entry, that peak retains its original m/z (i.e. not snapped to a > presumed interpretation) and a “?”. SpectraST seems to be smarter about not > assigning a silly interpretation. > > > > For the next peak with intensity 141.2, the old library had an > interpretation of a6^2 and the m/z is snapped to that interpretation. But > a6^2 is a super unlikely interpretation, or maybe let’s call it impossible > and wrong. In the new library, this same peak has been interpreted as an > internal fragmentation ion and the m/z has been snapped to that. The m4:6 > means that the peptide was fragmented in two places such that you get a > little ion that is TII and has a charge so you see it. Internal > fragmentation turns out to be fairly common and so m4:6 is far more > plausible an interpretation than a6^2. Same for m3:5 (ETI). Not necessarily > correct, but far more plausible than an a6^2 and a b6^2. > > > > So I’m guessing that the first spectrum comes from a version where > internal fragmentation was not yet supported, and the second spectrum comes > from a newer version of SpectraST that knows about internal fragmentation > and interprets peaks better. > > > > Does that seem like a reasonable explanation of what’s going on? Henry > might know more. > > > > Regards, > > Eric > > > > > > > > > > > > > > *From:* [email protected] <[email protected]> *On > Behalf Of *Kristina Plate > *Sent:* Friday, September 20, 2019 6:22 AM > *To:* spctools-discuss <[email protected]> > *Subject:* [spctools-discuss] Combination of consensus libraries and > import in MRM format via spectraST > > > > Hello again, > > > > as I described before (topic: *What happens during consensus building?*) > I have a problem with some messed up spectra in my spectral libraries. > > I combined four spectral libraries and built consensus spectra. Afterwards > I reimported the combined consensus spectral library in MRM format and > converted it in an assay library. When I checked the SPTXT file I found > some *slightly changed spectra* and new *"ion types" m*. Please see my > example of an old, normal spectrum and the corresponding new one below: > > > > Name: DIETIIQK/2 *(old entry, original library)* > > LibID: 4160 > > MW: 960.5481 > > PrecursorMZ: 480.2740 > > Status: Normal > > FullName: X.DIETIIQK.X/2 (CID-QTOF) > > Comment: AvePrecursorMz=480.5611 BinaryFileOffset=23385298 > FracUnassigned=0.09,1/5;0.23,11/20;0.06,7/31 MassDiff=-0.0014 Mods=0 NAA=8 > NISTProtein=Id_02 NMC=0 NTT=1 Nreps=1/1 OrigMaxIntensity=11 Parent=480.274 > Pep=Tryptic PrecursorIntensity=0 Prob=1.0000 Protein=1/Id_02 > RawSpectrum=file_02_04.69683.69683 RetentionTime=2745.2,2745.2,2745.2 > Sample=1/sample_02,1,1 Se=1^C1:pb=1.0000/0,fv=2.7040/0 Spec=CREATE > TotalIonCurrent=8.9e+003 iRT=33.4,33.4,33.4 > > NumPeaks: 31 > > 308.1581 226.5 ? > > 309.0826 359.4 ? > > 325.6734 1596.8 b6-35^2/0.000,b6-36^2/0.492 > > 329.1945 141.2 a6^2/0.000 > > 343.1920 668.4 > b6^2/0.000,y3-45/-0.042,y6-45^2/-0.516,y6-46^2/-0.024 > > 353.2183 967.7 y3-35/0.000 > > 354.2183 480.9 y3-35i/0.000 > > 355.2183 481.6 y3-35i/0.000 > > 358.1609 442.5 b3/0.000,y6-17^2/0.456,y6-18^2/0.948 > > 358.1970 96.7 ? > > 365.2030 293.0 ? > > 373.2367 296.2 ? > > 388.2554 1744.6 y3/0.000,b7-36^2/-0.955 > > 424.1714 315.8 b4-35/0.000,b4-36/0.984 > > 442.1820 644.1 b4-17/0.000,b4-18/0.984 > > 457.7633 546.4 p-45^2/0.000,p-46^2/0.492,y4-44/0.414 > > 458.2791 109.5 p-44^2/0.000,b4/-0.929 > > 471.7608 110.9 p-17^2/0.000,p-18^2/0.492 > > 472.2608 111.1 p-17^2i/0.000 > > 477.1321 111.6 ? > > 480.2740 336.0 p^2/0.000 > > 484.3130 224.9 y4-17/0.000,y4-18/0.984 > > 501.3395 523.5 y4/0.000 > > 553.1674 240.4 ? > > 567.3501 487.3 y5-35/0.000 > > 584.3766 247.1 y5-18/0.000 > > 602.3872 2935.1 y5/0.000 > > 696.3927 205.5 y6-35/0.000,y6-36/0.984 > > 714.4032 2211.4 y6-17/0.000,y6-18/0.984 > > 731.4298 10000.0 y6/0.000 > > 732.4298 414.9 y6i/0.000 > > > > Name: DIETIIQK/2 *(new entry, combined library)* > > LibID: 4673 > > MW: 960.5481 > > PrecursorMZ: 480.2740 > > Status: Normal > > FullName: X.DIETIIQK.X/2 (CID-QTOF) > > Comment: AvePrecursorMz=480.5611 BinaryFileOffset=20613807 > FracUnassigned=0.09,1/5;0.25,13/20;0.17,13/31 MassDiff=-0.0014 Mods=0 NAA=8 > NISTProtein=Id_02 NMC=0 NTT=1 Nreps=1/1 OrigMaxIntensity=11 Parent=480.274 > Pep=Tryptic PepContext=1/NTPR_DIHE PrecursorIntensity=0 Prob=1.0000 > Protein=1/Id_02 RawSpectrum=file_02_04.69683.69683 > RetentionTime=2745.2,2745.2,2745.2 Sample=1/sample_02,1,1 > Se=1^C1:pb=1.0000/0,fv=2.7040/0 Spec=CREATE TotalIonCurrent=8.9e+003 > iRT=33.4,33.4,33.4 > > NumPeaks: 31 > > 308.1581 226.5 ? > > 309.0826 359.4 ? > > 326.1680 1596.8 ? > > 328.2231 141.2 m4:6/0.000 > > 344.1816 668.4 m3:5/0.000 > > 353.2183 967.7 y3-35/0.000 > > 354.2183 480.9 y3-35i/0.000 > > 355.2183 481.6 y3-35i/0.000,m5:7/-0.016 > > 358.1609 442.5 b3/0.000 > > 358.1970 96.7 ? > > 365.2030 293.0 ? > > 373.2367 296.2 ? > > 388.2554 1744.6 y3/0.000 > > 424.1714 315.8 b4-35/0.000 > > 441.1980 644.1 b4-18/0.000 > > 457.2713 546.4 p-46^2/0.000,m3:6/0.006 > > 459.2086 109.5 b4/0.000 > > 471.2688 110.9 p-18^2/0.000 > > 472.2512 111.1 ? > > 477.1321 111.6 ? > > 480.2410 336.0 ? > > 484.3130 224.9 y4-17/0.000 > > 501.3395 523.5 y4/0.000 > > 553.1674 240.4 ? > > 568.3124 487.3 ? > > 584.2763 247.1 ? > > 602.3872 2935.1 y5/0.000 > > 695.4087 205.5 y6-36/0.000 > > 713.4192 2211.4 y6-18/0.000 > > 731.4298 10000.0 y6/0.000 > > 732.4675 414.9 ? > > > > I assumed before, that it was an issue with the consensus building, but I > figured out that it took action during the *import via spectraST into MRM > format*. > > > > Now, I just saw that I don't have this problem when I use the single > libraries. Even with consensus libraries built with with an older spectraST > version the MRM import works just fine. > > Maybe there is the *problem in the combination of the spectral libraries?* > I already got them as single consensus libraries and combined them with the > following command > > > > spectrast -cNcombined_lib -cd -cu -cr1 -cy1 -cAC -cJU > Lib_01.splib Lib_01.splib Lib_01.splib Lib_01.splib > > > > Could this be an issue here. Do I have to rebuild all libraries before > combining them? > > > > I would be really happy about any suggestions. Thank you in advance! > > > > Kind regards, > > Kristina > > ____________________ > > Kristina Plate, MSc. Biomathematics > > > > University of Greifswald > > Center for Functional Genomics of Microbes > > Institute for Microbiology > > Department of Microbial Proteomics > > Felix-Hausdorff-Str. 8 > > 17489 Greifswald > > > > phone: +49 3834 420-5925 > > fax: +49 3834 420-5902 > > -- > You received this message because you are subscribed to the Google Groups > "spctools-discuss" group. > To unsubscribe from this group and stop receiving emails from it, send an > email to [email protected]. > To view this discussion on the web visit > https://groups.google.com/d/msgid/spctools-discuss/1573a2bc-1b4a-4ca7-bc88-75579028df95%40googlegroups.com > > <https://groups.google.com/d/msgid/spctools-discuss/1573a2bc-1b4a-4ca7-bc88-75579028df95%40googlegroups.com?utm_medium=email&utm_source=footer> > . > > -- > You received this message because you are subscribed to the Google Groups > "spctools-discuss" group. > To unsubscribe from this group and stop receiving emails from it, send an > email to [email protected] <javascript:>. > To view this discussion on the web visit > https://groups.google.com/d/msgid/spctools-discuss/d206982e-1892-494c-8762-0c519ae3b10c%40googlegroups.com > > <https://groups.google.com/d/msgid/spctools-discuss/d206982e-1892-494c-8762-0c519ae3b10c%40googlegroups.com?utm_medium=email&utm_source=footer> > . > -- You received this message because you are subscribed to the Google Groups "spctools-discuss" group. To unsubscribe from this group and stop receiving emails from it, send an email to [email protected]. To view this discussion on the web visit https://groups.google.com/d/msgid/spctools-discuss/36b5011c-8fe7-4bf0-b4ab-990ffe31eadd%40googlegroups.com.
