Dear Support Team and Skyline users,
I am working with Skyline version 3.6 on a Windows 10 system and have following issue: If I explore my library and want to import all peptides which are in my library, than 50% of my peptides will not be imported into my Skyline document since it has no icon in front. I checked the forum and there are already a lot of similar questions around but no solution posted there helped me. I really appreciate any help at this point!
description in more detail:
We are working with cross-linked peptides and do a linearization of those with an artificial lysine in between as cross-linker modification. We use a script to transform peptides in there linearized form and create an .ssl file afterwards according to the Skyline website, with recomandat columns. If we use this .ssl file and our mxML files(converted from raw using MSconvert), we are able to build a library in Skyline. Afterwards we explore our library and do see not many icon assigned to our peptides. If I hover over the peptide I can see all modification masses and they are all defined correctly in peptide settings. There are still unassined intense fragments but they are likely from the cross-linking area and can not be assigned due to the fact that our cross-links where linearized. Additionally I also saw spectra with unassigned intense fragments with an icon which could be imported so I am not sure where are the difference in this cases. I checked all peptide and transition settings, played with it but nothing changed. I will still loose half of my peptides. I also changed the cut-off score while building the library but also no changes. I tried to subdivide my library into smaller lists to separate peptides with similar sequences because I thought Skyline may not be able to distinguish between peptides with simlilar fragments (without specific fragments for each peptide) but still no changes.
I hope you have suggestions. Thank you for your time and solutions you can provide me.
See Complete zip file attached.
Best wishes
Fränze
PS: After updating to Skyline 3.6 my Library explorer crashes quite often although all storage places for files are the remained the same. And one last question: How does Skyline calculate the CV values over replica? Is it the standard diviation within replica divided by the mean? |
| |
| Brendan MacLean responded: |
2017-06-11 17:32 |
Hi Fränze,
Thanks for all the detail. I guess this just came at a bad time, and it has taken me this long to get to it. Sorry about that.
First, the problem really boils down to the fact that your "Oxidation (M)" modification is not marked "Variable". This was probably a much more common mistake before I made the "Variable" checkbox on by default. I know I did this to myself enough to be annoyed by it. So, finally realizing that many more modifications are considered variable than "static" or applying always, I made it the default.
If I open your document, and immediately edit the "Oxidation (M)" modification to make it variable, then when I open View > Spectral Libraries, all spectra match as expected.
We are actually working on making the reasons behind not matching spectra clearer by summarizing them in the tooltip when you hover over the peptide. This work is not done yet, and I have passed along this great case to the developer working on it. I definitely know it can be difficult to figure these out, and I myself needed to use a debugger to figure this one out. We will improve this in the near future.
I am not sure about the "Library explorer crashes", but if you were posting them to the Skyline website, I expect we fixed them. If not, then we would need a more thorough report to be able to address the issue.
Finally, yes, CV in Skyline is always the mean / stdev.
Thanks again for all the detail you supplied and the example file. In the end it made this very easy to review and address. Sorry for taking so long.
--Brendan |
| |
| fraenze responded: |
2017-09-27 03:23 |
Hi Brandan,
Sorry for the late reply, your message got somehow lost in my inbox.
I tried your suggestion with the oxidized methionine, but it didn't work out in my hands. Additionally I have unimportable peptides without the ox modification. I also attached a new Skyline zip file, maybe I am still doing something wrong. I also tried the new version of Skyline because I was hoping the import problem will disappear in the new version, but it is still a problem.
I highly appreciate any hint or solution from your side. Maybe I am also missleaded by some terms in Skyline: Just to confirm, the total peptide numbers in the spectral library explorer is the same number as the peptide number in the right lower corner in the document right? So if I have 885 peptides in my library and just 633 in my document, than I am missing some peptides, right?
Best wishes and thank you for your help,
Fränze |
| |
| fraenze responded: |
2017-09-27 03:28 |
Hi Brandan,
I somehow can not attach my zip file, because I always get an unexpected error. |
| |
| Brendan MacLean responded: |
2017-10-15 15:46 |
Hi Fränze,
You can post your file to http://skyline.ms/files.url and then post here again letting me know where it is. It may be that you are now trying to post too big a file to attach it directly to your support request.
Actually, we is listed as "Peptides 1 through 77 of 77 total" in the View > Spectral Library form are actually peptide precursors, i.e. a modified peptide sequence plus a charge state.
So, in a case like the one above, you get the following message when you click the Add All button in the library explorer:
"This operation will add 63 peptides, 77 precursors and 462 transitions to the document."
Instead of getting 77 peptides, you actually get 63 peptides with 77 precursors, since some of the spectra match the same peptides with different precursor charges.
I am going to need you to be more explicit than "I tried your suggestion with the oxidized methionine, but it didn't work out in my hands."
What do you mean? If you download the document you posted to the first request and change your "Oxidation (M)" modification to variable, and then you use View > Spectral Libraries to view your spectral library, do you not see all 77 rows in the list with a peptide icon? Do you not get the message above when you click the Add All button?
Clearly, we are not quite connecting yet. Just not sure why. I am willing to keep trying if you can give me more information on what you are seeing. Maybe post a screenshot or two with your next explanation.
--Brendan |
| |
| fraenze responded: |
2017-10-17 07:02 |
Hi Brendan,
The file I uploaded was an extract from a bigger experiment file. So I check whether the oxidation of Methionine was set to variable or not and indeed it was set to fixed instead of variable. I changed it and I could import some peptides but not all, 25 peptides could not be imported but luckily it has a smaller impact than I thought, because I miss understood the peptide meaning in the document compared to the library explorer. So in the end I could still work with the quantitative results.
I can try to upload the not importable peptides (precursors) again if you wish to see whats the problem there.
Thank you for your response
Best wishes
Fränze |
| |
| Brendan MacLean responded: |
2017-10-17 08:09 |
Hi Fränze,
Maybe you could try it with Skyline-daily, if you haven't already. The latest version provides new helpers in diagnosing issues like this. Though, I admit they didn't make the Oxidation (M) issue immediately obvious. If you find a single peptide in the Spectral Library Explorer list that is missing the peptide icon (that it the first clue that Skyline won't be able to add it to the document), when you hover the mouse over the peptide sequence, Skyline-daily should now frequently show in the tip that appears some detail on why it will not be able to put the peptide precursor into your document, which is generally caused by 1 of 2 issues:
1. The precursor m/z is out of range for the instrument or DIA isolation range.
2. The modified state cannot be achieved with the current settings.
In #2, the tip does not yet cover your case where the modified state cannot be achieved because of a static modification which is missing, but it will cover the case when a modification that is present lacks a definition, by showing the modification delta mass in red with a question mark. And #1 will also show up with red error text indicating why the m/z is out of range.
Give that a try and let me know if it provides you any extra information. It would certainly be disappointing if this new support failed to provide any new information for all the cases in your document.
Thanks for sharing your document. If you still can't figure out those last 25, I would certainly be interested to understand myself to help us improve the tip further to make these issues easier to understand.
--Brendan |
| |
| fraenze responded: |
2017-10-18 06:32 |
Hi Brendan,
if I hover over the peptide sequence than the tip appears and show me the peptide sequence with my modification mass, charge and the precursor mass, but there is nothing marked red or with a question mark. I am not really sure whats the issue with these peptides. I will try to attach a smaller version of my file.
Best wishes
Fränze |
| |
| fraenze responded: |
2017-10-18 06:42 |
o.k. I used http://skyline.ms/files.url to upload the file. The description is cross-linking_notimportable_precursors and the name of the file is HSA_BS3_I_-DMSO (Berlin) 01.08.16_new skyline_notimportable.sky.zip. |
| |
| Brendan MacLean responded: |
2017-10-18 19:35 |
I don't see a spectral library in the file you posted. Maybe since you removed all targets, if you chose to share minimized, it may have left out the library, since minimized will only keep spectra that are used in your Targets list. Maybe try again. You can also use Edit > Manage Results to remove all results. I can't see why the results would matter in this. But, be sure to share Complete to give me your full library. Maybe you could also post a screenshot of one of the peptides and its tip showing in the View > Spectral Libraries form. Thanks.
--Brendan |
| |
| fraenze responded: |
2017-10-19 03:19 |
Hi Brendan,
Now it should be the complete file: same name with the description cross-linking_notimportable_precursor_complete.
Best wishes
Fränze |
| |
|
|
not importable cross-linked peptides.sky.zip