compomics / peptide-shaker

Interpretation of proteomics identification results
http://compomics.github.io/projects/peptide-shaker.html
48 stars 19 forks source link

MSFragger search result visualization #535

Open Baopusr opened 6 months ago

Baopusr commented 6 months ago

Dear peptide-shaker team, I receive this error when loading MSFragger result files :

Sat Mar 09 23:03:26 CST 2024 Importing sequences from human_uniprot-proteome_UP000005640-2021.06.20.fasta. Sat Mar 09 23:03:27 CST 2024 Importing gene mappings. Sat Mar 09 23:03:31 CST 2024 Establishing local database connection. Sat Mar 09 23:03:31 CST 2024 Reading identification files. Sat Mar 09 23:03:31 CST 2024 Parsing interact-1CuB_A549_TRIS_2_2h_hplc_F1_20231119.pep.xml. Sat Mar 09 23:03:31 CST 2024 An error occurred while loading spectrum matches from 'interact-1CuB_A549_TRIS_2_2h_hplc_F1_20231119.pep.xml'. This file will be ignored. Error: java.lang.NullPointerException See resources/PeptideShaker.log for details. Sat Mar 09 23:03:31 CST 2024 Warning: The software used to generate interact-1CuB_A549_TRIS_2_2h_hplc_F1_20231119.pep.xml was not recognized by PeptideShaker. Please create an issue on the tool website and we will add support for the software used. github.com/compomics/peptide-shaker/issues Sat Mar 09 23:03:31 CST 2024 No identification results.

Sat Mar 09 23:03:31 CST 2024 Importing Data Canceled!

Maybe due to the result file of the MSFragger-Labile Mode is different? Thanks

Best Baopu

hbarsnes commented 6 months ago

Hi Baopu,

I'm afraid that pep.xml is not as standardized as one may think. We therefore have to implement individual support for the different search engines producing pep.xml files, for example to correctly parse the scores. As far as I remember, this is not something we have yet done for MSFragger. However, if you are able to share an example of the MSFragger data you are trying to load in PeptideShaker we can try to have a closer look at see whether it is something we can consider implementing.

Best regards, Harald

Baopusr commented 6 months ago

Dear peptide-shaker team,

Thank you very much! I'm sorry I don't know much about peptideshaker, but I'm amazed by the software's ability to parse search results. Thank you very much for your willingness to try to help solve my problem, and the following is my search results, its corresponding literature is "Enhancing Cysteine Chemoproteomic Coverage through Systematic Assessment of Click Chemistry Product Fragmentation". Looking forward to your reply~ result_Labile_20240312.zip result_NonLabile_20240312.zip

Best wishes, Baopu

hbarsnes commented 6 months ago

Hi Baopu,

I had a look at the files you shared plus the additional data from PRIDE. Adding support for loading MSFragger pep.xml files should not be too difficult, however there is an issue with the decoy annotation in the FASTA file used as it is not of the format that we support and I'm not sure how easy this would be to fix.

Furthermore, it is worth noting that PeptideShaker is not strictly a tool for visualizing search results but it also reprocesses and analyzes them. Hence, the results displayed in PeptideShaker will not be identical to the results in the original paper. With this in mind, you may therefore instead consider to reprocess the raw data from the paper via SearchGUI and then load those results in PeptideShaker. But this of course all depends on what you are trying to achieve my utilizing the given dataset.

Best regards, Harald

Baopusr commented 6 months ago

Dear peptide-shaker team,

Thank you very much for your reply and reminder. I did mistake PeptideShaker for a pure tool for visualizing and analyzing search results. However, this time I want to use the newly developed Labile PTM mode search mode of MSFragger. SearchGUI may not support this new type of search mode. But thank you very much for your help. If you are willing to introduce this new type of search mode to SearchGUI, I will be the first to try it out. Thank you!

Best wishes, Baopu

hbarsnes commented 6 months ago

Hi Baopu,

I'm afraid that we do not currently have the resources to implement this type of PTM search in our tools. I will however keep the issue open in case the situation changes in the future.

Best regards, Harald

Baopusr commented 6 months ago

Dear peptide-shaker team,

Thank you, I will continue to follow the latest developments of SearchGUI and PeptideShaker. The BioTools developed by Compomics have inspired me a lot!

Best wishes, Baopu