glycoinfo / GlycanBuilder2

7 stars 5 forks source link

Invalid GWS using Glycan.toString() #24

Open ReneRanzinger opened 3 years ago

ReneRanzinger commented 3 years ago

Example GlyTouCan ID: G17689DH

Using the Glycan.toString() method: redEnd--?b?D-b-D-Glcp,p((--2NAc)--4b1D-b-D-Glcp,p(--2NAc)--4b1D-b-D-Manp,p(--3a1D-a-D-Manp,p--2b1D-b-D-Glcp,p(--2NAc)--4b1D-b-D-Galp,p--3a2D-a-D-Neup,p--5NAc)--6a1D-a-D-Manp,p--2b1D-b-D-Glcp,p(--2NAc)--4b1D-b-D-Galp,p--3a2D-a-D-Neup,p--5NAc)--6a1L-a-L-Fucp,p$MONO,perMe,Na,0,redEnd

But it should be: freeEnd--?b1D-GlcNAc,p(--4b1D-GlcNAc,p--4b1D-Man,p(--3a1D-Man,p--2b1D-GlcNAc,p--4b1D-Gal,p--3a2D-NeuAc,p)--6a1D-Man,p--2b1D-GlcNAc,p--4b1D-Gal,p--3a2D-NeuAc,p)--6a1L-Fuc,p$MONO,perMe,Na,0,freeEnd

The upper string is no longer compatible with GWB and will not load. Main issues are the replication of anomer, ring and configuration in residue name that prevent a matching and the separation of 2NAc from monosaccharides.

e15d5605 commented 3 years ago

I tested a conversion functionality of GWS(GWB) format from WURCS using GlycanBuiler (1.12.0).

I also entered GWS into GlycanBuilder2.\ As a result, the following glycan image was depict.

Based on the above, I considered that this problem was resolved in the latest version of GlycanBuilder2 (1.12.0).

edwardsnj commented 3 years ago

Actually, the issue is (based on some experimenting just now), that if you use org.glycoinfo.application.glycanbuilder.converterWURCS2.WURCS2Parser and WURCS sequence to construct the glycan, you get correctly formatted GWB sequence, but if you use org.eurocarbdb.application.glycanbuilder.converterGlycoCT.GlycoCTCondensedParse and GlycoCT sequence to construct the glycan, you get incorrectly formatted GWB sequence.

The sequences I used were from here: https://glycans.glyomics.org/glycandata/G17689DH

Any insight?

Why would GlycoCT-based glycan generate different GWB format sequence?

Thanks,