Thanks a lot for your explanation and suggestions.
I think the problem may be associated with the option of "RT for scoring and filtering". In addition, the MS-DIAL internal library (lipidblast) need to be clicked the option (RT for scoring and filtering), right?
Another issue of "Accurate mass tolerance (MS1)" was also stranged.
I used to perform multiple lipidomics data with MS-DIAL and the value was set 0.01 in the option of "Accurate mass tolerance (MS1)" under the identification tab (please see the attached figure). Then it could obtain a lot of ref. matched lipids in final results.
However, the 0.025 was set for metabolomics could not match well. I will check the mass accuracy later.
I encounter an issue of the MSP database selection in metabolomics in MS-DIAL as follow,
I performed the plasma metabolomics test and followed the MS-DIAL paper (nature method, DOI:10.1038/NMETH.3393). The polar metabolites extraction and LC-MS conditions were similar to the paper description.
My MS data (Waters, Mse data) was processed by MS-DIAL and the MSP database file was chosen All Public (please see the attached figures). The ref. matched metabolites are around 400. However, the ref. matched metabolites are 0 when I chose the Fiehn HILIC database. (Because I used HILIC column and follow similar LC-gradient). Why is so large difference by using two databases and at least some metabolites should be overlap?
In addition, another important value is Accurate mass tolerance (MS1). When I put 0.025 Da (follow tutorial), the ref. matched metabolites are 0 but I put 0.05 Da, the ref. matched metabolites are around 400. I checked my MS raw data, the mass accuracy could be 0.025Da tolerance but not sure why the MS-DIAL can't recognize it well.
Any comments and suggestions are all welcome and hope the experienced user could help me solve them.
Hello, the issue wasn't resolved and I already used the data for further analysis. Fortunately, only fewer features were observed similar issues. I also tick the option of "gap-filling by compulsion", but it still has zero value.
I recently encountered a problem, some of the features in alignment results appeared a lot of zero values but the features could detect well in MS raw data (please see the attached file). Here we take a DAG (18:1/18:1) as an example, the mass of DAG (18:1/18:1) could be observed in all 6 replicated data but in final alignment list, M-1 and M-5 peak height changed obviously compared with other four data. From MS-Dial, it seemed to mislabel it as isotope (M+1) in M-1 and M-5.
So could anyone tell me what's the reason and how to correct it by setting the proper MS-DIAL parameters?
I already selected the function of retention time correction and now data is processing. Here I would like to know need I decrease the Rt time tolerance under the alignment parameters panel correspondingly? I chose 0.5min tolerance before.
My MS raw data were collected recently but it exhibited large retention time drift. My QC4 to QC 10 were distributed regularly in the sample list but expressed large retention time drift, as shown in the attached figure. The Rt decreased gradually as time passed.
So could MS-DIAL correct the retention time drift and how to set the alignment parameters correctly?
Hope the users have similar experience could give me some useful suggestions. Thank you very much.
Thank you for your suggestion and it does work. we could find the spike standards in the final alignments list.
But one more problem is the processing time. Now I total have six raw data but data process was very time-consuming.
My raw data was MSE data and from Waters MS machine. I remember the MSMS-abundance cut off should set a higher value to avoid a long time process. Now my parameter was set 500 as previously attached figure but also need 3-4 hours to finish total 6 data process.
So did u have any experience in this issue and could you reply to me in your available time?
I recently processed my raw data (waters Q-Tof) using MS-DIAL 4.16. But I encounter some problems as follow,
1. My MS data exhibited a little large mass accuracy, I totally had 14 samples but the mass window should be +- 0.04da during all samples (1-2 sample had large mass variation).
2. So if I set the MS1 tolerance as 0.02 under mass accuracy in data collection of the MS-DIAL panel, some spike standards in these large mass variation samples have no peak height in the final alignment results list (please see the attached figure 1).
3. So I set MS1 tolerance as 0.035 and 0.05, respectively. It could find the correct peak height of these spike standards in the peak list including large mass variation samples. However, the spike standards mass features never be in the alignment results even I modified my alignment parameters (please see attached figure2 ).
4. I am not sure whether have other users encounter a similar issue ( spike standards in peak list but not appear in alignment list).
Please give me some suggestions or solutions in your available time.
Thank you for your answer in details, you provide many useful information and suggestion. Actually, the XCMS optimized experiments was finished, and I will try other methods you provided to compare. The final aim is find real biological information from raw data, so thank you for you help again, I hope could consult with you profession questions in the future.
I believe that many people encounter such kind of question, XCMS and Progenesis QI, two kind of softwares who is better or more accuracy? Now I have a project, it contains many cell samples, the samples were analyzed in waters synapt G2 Si and the raw data was converted into the netCDF format by databridge software. Then the CDF files were processing by XCMS and QI, Finally, the table list had a large difference. the table list of XCMS is about 17000 features and the QI only is about 7000 features. So I am not sure who is better ? Maybe many nosie features go into the XCMS table list or maybe the QI lost the important features? Base on the example, I hope we may discuss the data processing workflow , how to ensure that the real difference existing in the raw data could be dig accurately by our data processing step ? If necessary, I can upload my R scripts command for everybody check. Thank you
