TCB-yehong
commented
1 year ago Hi Jun, Thank you for sharing the error messages! The error seems to be involving R package version issues. I have modified the pipeline so it's performing correctly in the latest R (at this time), which is version 4.2.2. I have updated the description to say 4.2.2 instead of 4.0.3. Could you please install version 4.2.2 and please feel free to delete 4.0.3.
Yes the input format you have might be the major problem, I think I have received an email from you, which I have replied just now, but I will also include the suggestions here: Regarding your question, yes I'm afraid there will have to be reformatting of the input data. Different spectral analysis software could output the data in different ways, that might be why you have the different data format, with multiple phosphorylation sites marked on the same row. To change from this format to the PhosPiR preferred input format, there's quite a bit of work to be done. First, for each site, you have to check if it exists in multiple rows (sometimes the spectral analysis software output the same peptide on separate rows when the peptide has different number/position of phosphorylations). If it does, the intensity of this site should be combined. Then, for each site, you have to center it, which means, you would need 15 sequences before and after this site, or at least 7 sequences before and after this site. The final input would have the following format: each row would have 1 site centered, with 15 sequences (or at least 7 sequences) before and after it. The sample columns for this row would display the intensity values of this site, after the intensity have been combined from different rows that includes this particular site (if only one row includes this site, then this row's intensity can be used). For the rows with multiple sites in your data, this should be done for every site, and each site should take a new row. For the rows with a single site, the same thing should be done too, which means the site should be centered, and it should be checked with other rows to see if the same site is included elsewhere. After you have the new rows, the old rows can be removed. This might take a lot of time to do, one thing that might make this faster/more automated, is that if you have the raw data, you can run it through MaxQuant software, and the output will have the format that I described, and can be input directly to PhosPiR. Please let us know if you have further questions about how to proceed!
The input format will have to be changed in order for PhosPiR to not give an error for kinase analysis. However, if you would just like to analyze the intensities for the phosphopeptides without kinase analysis, you could exclude phosphosite identity entirely and treat the dataset like a proteomics dataset, that way you could still analyze your data with PhosPiR, just without the kinase analysis. Hope that helps!
junyang1966
janeseto
Hi,
Thanks for developing the software PhosPiR. I really would like to try it on our phosphoproteomic data. I don' have much experience in programming as well as R. I watched the two YouTube videos. So I downloaded and installed R 4.0.3 using the link in the Readme file. I downloaded the PhosPiR pipeline exactly as described in the Readme file. I put the Run file in the R window. I did all these in my PC computer. However, I ran into a series of problem not mentioned in the YouTube video. Please see the attached file of what was shown in my R window. I am not sure whether it is a big or small problem and how to solve it.
Another thing is that I have multiple phosphorylated sites on single peptides. By reading the thread of one issue, I think the problem was addressed. However, due to my limited knowledge in bioinformatics and computer language, I don't understand the solution. At this point, I would like to know how to prepare the input file for multiple phosphorylated sites and what the format should be.
Many thanks in advance. Jun GraphAppPrintJob.pdf