Spectrast: peptides with modifications are removed when converting NIST .msp formatted library to .splib.

42 views
Skip to first unread message

Ray

unread,
May 27, 2020, 3:48:01 PM5/27/20
to spctools-discuss
Hi,

I was trying to convert the NIST consensus human hcd library (.msp format) to splib format with spectrast built from TPP 5.2.

After running the command:
spectrast -cNhuman_hcd human_hcd_tryp_best.msp

The resulting human_hcd.sptxt file contains significantly less spectra (273,137) than the original human_hcd_tryp_best.msp file (398,373). It seems that all peptides with modifications are discarded.

For example, in human_hcd_tryp_best.msp file, there is an entry: AAAAAAVGPGAGGAGSAVPGGAGPCATVSVFPGAR/3_2(0,A,Acetyl)(24,C,CAM)

But there is not such peptide in the output human_hcd.sptxt file. Did I miss some command-line arguments when converting the library?

Thank you.

Ray

Wenguang

unread,
May 28, 2020, 3:10:05 PM5/28/20
to spctools-discuss
Hi Ray,

Maybe try with -M option, which allows you to define customized modification types. You can find more details about -M option from SpectraST wiki.

Best,
Wenguang

Henry Lam

unread,
Jun 3, 2020, 3:49:28 AM6/3/20
to spctools-discuss
Ray,

This is because SpectraST does not recognize this way of specifying modification in "new" NIST files.

If you are comfortable compiling your own TPP, try downloading SpectraSTMspLibImporter.cpp from the latest "trunk" version of TPP, replace and re-compile. This should fix the problem. 

Henry

Henry Lam

unread,
Jun 3, 2020, 4:01:24 AM6/3/20
to spctools-discuss
Ray,

Sorry -- you also need to replace:

Peptide.cpp
Peptide.hpp
FileUtils.cpp
FileUtils.hpp

Hopefully that should work. Please let me know if it still doesn't.

Henry
Reply all
Reply to author
Forward
0 new messages