Inquiry Regarding MSstatsPTM Usage and Normalization Methodology

[曹高祥]

Dear all，

In the label-free quantification workflow, I understand that the msstats_format_lf file contains lists of both modified peptides and proteins. These protein lists seem to be derived directly from the results of the current experiment(evidence file). If I were to conduct a separate protein quantification experiment, could I generate a protein list using MSstats or MSstatsTMT? And subsequently, could I replace the protein list generated by MSstatsPTM with this newly generated protein list, enabling me to proceed with the downstream analysis? Does anyone have any good methods and code to share?

Best,

Gaoxiang

[Anthony Wu]

Hi,

I'm not sure if I understand your question correctly.  Are you looking to create an initial MSstatsPTM format input using one of the converters, then replace the PROTEIN variable with another msstats_input list from a different experiment (possibly with results from a different upstream tool)?  

For example, are you thinking of a workflow looking something similar to below:

```

msstats_data = FragPipetoMSstatsPTMFormat(fragpipe_input, fragpipe_annotation, fragpipe_input_protein, fragpipe_annotation_protein, label_type="TMT", mod_id_col = "STY", localization_cutoff=.75, remove_unlocalized_peptides=TRUE)

maxq_tmt <- MaxQtoMSstatsTMTFormat(evidence, proteinGroups, annotation.mq)

msstats_data$PROTEIN = maxq_tmt

# Run data summarization and group comparison on msstats_data

```

Thanks,

Tony

[曹高祥]

Hello, Tony,

 

I'm sorry I didn't clarify my question earlier. Here is my question: I conducted two independent mass spectrometry experiments for quantifying methylation modifications. One experiment used a label free quantification strategy, while the other experiment involved quantification of methylation modifications based on enrichment strategy. I performed two separate searches using the MaxQuant software. I would like to know how to normalize protein data for methylation modifications in the context of label free quantification proteomics data using MSstatsPTM?

 

Thank you.

[Anthony Wu]

Hi,

To normalize your data for label free quantification proteomics data using MSstatsPTM, you can use the `normalization` and `normalization.PTM` parameters.  `normalization` corresponds to the first experiment that uses label free quantification strategy and `normalization.PTM` corresponds to the second experiment that performs quantification of methylation modifications based on enrichment strategy.  This link should provide more information on those parameters.  

The example below represents equalizing medians normalization for both experiments:

summarized_data = dataSummarizationPTM( data, normalization = "equalizeMedians", normalization.PTM = "equalizeMedians", )

Using the output from the previous code, you can run the below code to validate if normalization was applied correctly (see link for details):

dataProcessPlotsPTM( summarized_data, type = "QCPlot", )

Please let me know if you have any more questions or if I misinterpreted your question.

Thanks,

Tony

[曹高祥]

Hi，

I'm glad to receive your response, and I have benefited a lot from it. However, I have some other questions. Since my data comes from two separate searches, they are two different datasets. How should I proceed with the next steps using MSstatsPTM, given that the data typically used in MSstatsPTM is from a single search result?

Thanks，

Gaoxaing

[Anthony Wu]

Hi,

I'm a little confused from your statement that "the data typically used in MSstatsPTM is from a single search result".  

I believe the data used in MSstatsPTM is typically from two separate searches.  If you look at Fig. 3 of this paper on MSstatsPTM, you can see that identification and quantification for the first and second experiments happen separately and independently of each other.  

So given you have 2 datasets, we can use the MaxQuant converter function to take your 2 datasets and transform it into msstatsPTM format.  Pay attention to the two parameters `evidence_prot` and `annotation_protein`, which is where you input your dataset from the first label free experiment.  `evidence` and `annotation` parameters are used for your enriched dataset.

msstatsptm_format = MaxQtoMSstatsPTMFormat(evidence=dataset_enriched, annotation=dataset_enriched_annotation,

evidence_prot=dataset_label_free,

annotation_protein=dataset_label_free_annotation, fasta_path="fasta_file_from_uniprot.fasta")

Let me know if you have any more questions or if I didn't interpret your question correctly.

Thanks,

Tony