Viewing non-crosslinked peptides?

[Alemarcohen]

Hi Xi people!

I just run a search using the xiSearch/xiFDR/xiView pipeline.
Is there a way to see the NON cross linked peptides identified in the workflow (either in xiView or in any of the generated .xls files)? I'm 'validating' the results obtained with ProteomeDiscoverer and want to compare the data of the 'regular' peptides obtained in both searches.

Thanks!

Alex

[lutzfischer]

xiSEARCH returns (unless instructed not to) also linear peptides.
Also xiFDR also does FDR on linear peptides and reports these (files Linear_PSM, Linear_Peptides). However unless you defined a FDR-cutoff fr peptide pairs or PSMs these will be unfiltered. But the files contain a FDR column. So you can just filter these on your own.

One thing to keep in mind is that xiSEARCH by nature of doing a crosslink search is likely to return fewer linear matches for a given confidence then a search just for linear peptides. The reason being that the cross-linked search space is much (much) larger then the linear one. So some of the correct linear PSMs will be "covert" up by false crosslink PSMs. I.e. the correct linear match will sometimes have a lower score then an incorrect random crosslinked match and then the crosslink will be reported.

On the other hand the linear search will match some crosslinked spectra incorrectly as linears. So there is a bit of an balance.

[Alemarcohen]

Thanks Lutz! I found it.... thanks for your patience, I spend more time running the lab so my 'bioinformatics' is rusty, I'm more of a GUI person, that's why I amazed by your XL pipeline. I tried others, and mostly failed in the implementation due to the lack of file/data/software incompatibility and support.
Quick followup... do you have any GUI to visualize (and mine/sort) all the .csv files? I dont think XiView has an option to show this data, correct? As a CORE facility, I need to report the data in a format that is non-MS researcher friendly.
Thanks again!

[lutzfischer]

xiVIEW can be used to look at the network. If you want to look at the linear matches you can use xiSPEC (https://spectrumviewer.org/) to look at the linear only spectra (or all spectra for that matter).
As for mining these we don't have a gui at the moment. What do you want to look for? Maybe we can come up with something ...

[Alemarcohen]

Thanks Lutz I was caught in other projects! back to my xl samples now. All I thought for the linear peptides was for example, when you clicked on the nodes (proteins) in the XiView, an option where you could also see the linear peptides which were identified for that protein, including the modified monolinks. That would be cool to compare with the hits obtained by 'typical' proteomics programs (eg, ProteomeDiscoverer) for linear peptides without having to open the .xlsx file produced by XiSearch with is not friendly to browse. Cheers!!! Alex

…

On Thu, Sep 12, 2019 at 1:45 PM Lutz Fischer ***@***.***> wrote: xiVIEW can be used to look at the network. If you want to look at the linear matches you can use xiSPEC (https://spectrumviewer.org/) to look at the linear only spectra (or all spectra for that matter). As for mining these we don't have a gui at the moment. What do you want to look for? Maybe we can come up with something ... — You are receiving this because you authored the thread. Reply to this email directly, view it on GitHub <#43?email_source=notifications&email_token=AHDLPAZDEBDNCUQQ22IYKFTQJJW3LA5CNFSM4IUAAD4KYY3PNVWWK3TUL52HS4DFVREXG43VMVBW63LNMVXHJKTDN5WW2ZLOORPWSZGOD6SQUDY#issuecomment-530909711>, or mute the thread <https://github.com/notifications/unsubscribe-auth/AHDLPA24SHNFTZZ26ZESHRDQJJW3LANCNFSM4IUAAD4A> .

[colin-combe]

Hi Alex,

All I thought for the linear peptides was for example, when you clicked on the
nodes (proteins) in the XiView, an option where you could also see the
linear peptides which were identified for that protein, including the
modified monolinks. That would be cool

Indeed, that would be cool.

In the case of monolinks, you can do this by uploading your data to v1 of xiNET (http://crosslinkviewer.org/CLMS-CSV.php#productTypes , monolinks called "linker modified peptides" there).

Unfortunately, this functionality doesn't exist in xiView as yet (Rappsilber-Laboratory/build-xiview#19).

Showing the linear peptides in the expanded bars in xiNET is also a good idea. 'Showing linears in the network' or 'showing sequence coverage' has come up before, its a bit like Rappsilber-Laboratory/xiSPEC_website#18

If you upload a combination of crosslinked and linear peptides to xiView at the moment then it only loads the crosslinked ones into the interface (Rappsilber-Laboratory/build-xiview#7).

I'll have a look at what can be done to improve the situation and get back to you,
cheers,
Colin