Re: Call for comments: GlycoCT issues

已查看 3 次
跳至第一个未读帖子

kki...@soka.ac.jp

未读,
2019年1月29日 21:12:112019/1/29
收件人 glycoinformatics...@googlegroups.com、iss...@noguchi.or.jp
Dear all,

Thank you for your comments.  It seems that we’re in agreement in general.
The only issue seems to be the ordering of residues when representing a monosaccharide
composition in GlycoCT.  If there is documentation of some kind or if anyone has
an idea of how they think the residues in the RES section should be ordered in a 
composition, please add a comment under the Compositions section in the Google Doc.


Thank you!

Kiyoko

P.S. I just realized that it’s not the end of the month yet :)  So feel free to comment about anything
else by February 8th as well.  

------------------------------------------------------
Kiyoko F. Aoki-Kinoshita
Professor
Faculty of Science and Engineering, Soka University
------------------------------------------------------






On Jan 17, 2019, at 18:19, 木下 聖子フローラ <kki...@soka.ac.jp> wrote:

Dear GLIC members,

During our handling of GlycoCT in GlyTouCan, we have noticed some discrepancies in the way GlycoCT is written by various database and software developers.  Issaku has summarized the main points that we would like to confirm with all of you in the Google document whose link is below. We plan on developing a GlycoCT validator which will check these points among others. Please take a look at this document, as we would like to make sure that there is a consensus.  Please make any comments directly in this document by the end of the month.  Once everyone seems satisfied with these points, we will implement the validator from mid-February.


Thank you in advance,

Kiyoko and Issaku

------------------------------------------------------
Kiyoko F. Aoki-Kinoshita
Professor
Faculty of Science and Engineering, Soka University
------------------------------------------------------







回复全部
回复作者
转发
0 个新帖子