Hi, I am testing MS-DIAL v 5.1.230222 and I found a problem with "Gap filling by compulsion" option.
Quote
Gap filling by compulsion: If you check this, the peak recognition is performed by the average peak width of samples having the metabolite feature even though no local maximum is observed in the chromatogram. This is validated by default.
I do not want it, I switched it OFF. However, there is peak intensity and peak area value even in case of BLANK, where there is no peak / data at all (see attached image). It should not be there. O.
Hi, I processed an identical set of metabolomics data using identical parameters either in ver 4.9.221218 or ver 5.1.230222.
MS-DIAL 4.9.221218 processed 164 files in 7 minutes (from the first "dcl" file to last) and saved the alignment file in next 6 minutes. Total 13 minutes.
MS-DIAL 5.1.230222 processed 164 files in 20 minutes (from the first to the last) and I canceled the alignment after 60 minutes (CPU at constant ~6.6% and slowly increasing allocated memory ~14.2 GB). Total 20 + 60...
Update, after fresh restart: MS-DIAL 5.1.230222 processed 164 files in 19 minutes (from the first to the last) and finished the alignment in 2 hours and 11 minutes. Total 2 h 20 min. O.
Hi, The computer has 8 CPU cores, thus 16 threads are available. MS-DIAL 4.92 is fine and uses 16 threads. MS-DIAL 5.1.230222 uses max 8 threads. Why there is this limitation? Thanks, O.
Dear Hiroshi, could you please unify the LipidBlast nomenclature with the recommended format? (DOI 10.1194/jlr.S120001025, Liebisch et al. JLR, 2020)
I am using MS-DIAL v4.90 (Windows) and the main problem is with the reported oxygens. Using LM notation the 'position' should be before the "O" and the number of oxygens after the "O".
MS-DIAL name is: PI 17:0_18:2;2O The normalized name at this level should be: PI 17:0_18:2;O2
Mrmprobs Console App requires the following args: MrmprobsConsoleApp.exe <analysisType> -i <input folder> -o <output folder> -m <parameter file> -r <reference library file> -d <dia dictionary file (required in abfdia)> Where: <analysisType> is one of abfmrm, mzmlmrm, mzmlscan, abfscan, abfdia (required) <input folder> is the folder containing the files to be processed (required) <output folder> is the folder to save results (required) <method file> is a file holding processing properties (required) <reference library file> is a file containing rt/transition/ratio/range information at tab-delimited text. <dia dictionary file> is required in DIA-data process containing SCAN/DIA ScanID MzBegin MzEnd information as tab-delimited text
Hi Hiroshi, I found a possible bug. I have one peak close to my analyte RT. When my analyte is not present, MRMPROBS selects the wrong peak nearby. As expected.
Now I want to manually remove the wrong peak. If I do it in the "View" mode using left and/or right mouse button, raw data matrix shows the value or N.D. Fine.
If I do it in the "Edit" mode and I select no peak, raw data matrix still shows the previous (last) value. And it should be N.D. as above. Best, Ondrej
Hi, what is the problem? It looks fine. You can probably play with deconvolution params in MS-DIAL to get cleaner spectra and more reasonable MRMs. Ondrej
Hi Yuki, thank you for the alpha version. I have a blank + 3 abf files, lcmsdda.
Win GUI MS-DIAL 4.36 works fine
Win console MS-DIAL 4.36 works only with tutorial param file. If I export the params from v4.36, it doesn't work.
Linux console alpha does not work. Log is pasted below.
. Problems
Paths. It seems that linux app doesn't parse the paths correctly. I had to use absolute paths to run it and the output is in the root instead of the output folder. Also the file names contain path (see ls output below)
Blank - when I set a blank filtering, I can't set blank file. I assume this was not implemented yet.
MSPs - I can't upload my MSP. I assume only included libraries are supported now.
Params - I have to use a simple param file from the demo files or edit the full version exported from MS-DIAL. Most probably MSP and 'blank' definition problem.
ls -alF total 60548 drwxrwxr-x 2 ondrej ondrej 4096 říj 30 09:54 ./ drwxrwxr-x 4 ondrej ondrej 24576 říj 30 09:54 ../ -rw-rw-r-- 1 ondrej ondrej 738521 říj 30 09:54 '\AlignResult-20201030953.arf2' -rw-rw-r-- 1 ondrej ondrej 374618 říj 30 09:54 '\AlignResult-20201030953.dcl' -rw-rw-r-- 1 ondrej ondrej 19111850 říj 30 09:54 '\AlignResult-20201030953.EIC.aef' -rw-rw-r-- 1 ondrej ondrej 6066618 srp 25 09:38 blank.abf -rw-rw-r-- 1 ondrej ondrej 4704 zář 17 07:38 LIBneg_FAs.msp -rw-rw-r-- 1 ondrej ondrej 1535 srp 21 2017 Msdial-lcms-dda-Param.txt -rw-rw-r-- 1 ondrej ondrej 4819494 říj 30 09:54 '\project_2020_10_30_9_53_50_tmpFolder\peaklist_0.pll' -rw-rw-r-- 1 ondrej ondrej 4929846 říj 30 09:54 '\project_2020_10_30_9_53_50_tmpFolder\peaklist_1.pll' -rw-rw-r-- 1 ondrej ondrej 4917734 říj 30 09:54 '\project_2020_10_30_9_53_50_tmpFolder\peaklist_2.pll' -rw-rw-r-- 1 ondrej ondrej 4923190 říj 30 09:54 '\project_2020_10_30_9_53_50_tmpFolder\peaklist_3.pll' -rw-rw-r-- 1 ondrej ondrej 8745 říj 30 09:54 Project-20201030953.mtd2 -rw-rw-r-- 1 ondrej ondrej 5313579 srp 25 09:38 S1.abf -rw-rw-r-- 1 ondrej ondrej 5509016 srp 25 09:38 S2.abf -rw-rw-r-- 1 ondrej ondrej 5222554 srp 25 09:38 S3.abf
Hi, it was created for targeted (MRM) analysis. You can run the statistics and quant in excel or use the implemented internal standard approach. Scan or DIA mode allows you to dissect 'raw' data from the file. We use it for pseudoMRMs (parent > parent) and you can review all samples in parallel (similar to Skyline). Best, Ondrej
Hi Hiroshi, I found a problem in MRMPROBS 2.50 when the peak is very high. With some peaks, we reach the limit of 32-bit integer (ranges from -2,147,483,648 to +2,147,483,647). I see the peak height well, but when I change the info to peak area, I get a negative value. Thus, when exported as raw matrix, I get zero values. Could you please change it to 64 bit integer? Thank you. Ondrej
Hi Hiroshi, thank you. It works now. Tested on two different projects, 25k and 35k peak spots from 350 samples. The bar grows slowly and then jumps several times till 100%. Ondrej
Hi Hiroshi, I have a problem with "Show ion table" function. I am not able to use it in some projects. Let's say I have ~800 peaks spots (Blank+, MS2 acquired+, Suggested+) that I need to review. When I click on the icon, the progress bar "Progress of preparing alignment spot table viewer" stops at random position and never reaches the end. Sometimes is freezes at ~20%, sometimes at ~80%. I can use the X to close the bar and try again. HW resources are available. If I completely rebuild the project, it helps sometimes, but sometimes it freezes again. True for multiple versions of MS-DIAL. Could you please suggest how to find what is wrong? Thank you, O.
The TMP folder within the project folder. Usually something like project_2020_9_9_12_33_13_tmpFolder ... peaklist_101.pll peaklist_100.pll peaklist_10.pll peaklist_1.pll peaklist_0.pll ...