search

mobiusklein / glycresoft

An LC-MS/MS glycan and glycopeptide search engine
https://mobiusklein.github.io/glycresoft/
Apache License 2.0
8 stars 7 forks source link

How to add "Methyl" group and "Acetyl" group " into database? #10

Closed zk3353619 closed 3 years ago

zk3353619 commented 3 years ago

Hi Joshua, When I was "Building a Combinatorial Glycan Hypothesis", it's difficult to add the "Methyl" group and "Acetyl" group into the hypothesis---Is there a method to add these two compositions into database? image

mobiusklein commented 3 years ago

To add floating substituents to a glycan composition, you need to prefix their names with an @ and write them lower-case. The names @acetyl and @methyl should work. You can add arbitrary named chemical compositions with the following notation: #<name>#<formula>.

In my email notification, I saw another message from you about adding an mzML file to a project, but I don't see it on GitHub itself. Was that issue removed?

zk3353619 commented 3 years ago

Thank you so much for your quick reply! Your advices worked successfully. And the second question has been solved.

Here is another issue I want to discussed with you: Now I want to use the GlycReSoft to analyze the LC-MS data from plant polysaccharide: 1). Do you have any suggestions or a advices for this? 2). Is this analysis appropriate? As we know the GlycResoft was designed for N-glycan.

mobiusklein commented 3 years ago

Plant polysaccharides as in plant N-glycans? It's certainly doable. Plants have different glycan biosynthesis pathways, but the chitobiose core is still there. You can use this combinatorial hypothesis:

Hex 2 7
HexNAc 2 6
dHex 0 3
Xyl 0 1

HexNAc >= dHex

If you're trying to analyze the really long mostly linear polysaccharide chains, it is a bit harder. If you have a specific taxonomic ID, you could use the command line client to query glySpace for glycans belonging to a specific taxonomic tree: glycresoft build-hypothesis glycan-glyspace -t <taxon-id> -i path/to/file.db though this is liable to not produce a comprehensive database either.

If you're familiar with Python, you can write an arbitrary program in Python using glypy to build a hypothesis text file with one glycan composition per line of a file. Similarly, you could download a curated list from https://glyconnect.expasy.org/browser/taxonomy and use glypy.io.glyconnect to parse it into the appropriate format.

mobiusklein commented 3 years ago

Did this resolve the problem you were facing?