The latest versions of COOT have a nifty carbohydrate building module for N-linked glycans, and Buccaneer can do carbohydrate validation ( see appropriate manuals ).
This article also highlights some of the issues and pitfalls you might encounter. https://www.ncbi.nlm.nih.gov/<https://www.ncbi.nlm.nih.gov/pmc/articles/PMC5297920/>pmc<https://www.ncbi.nlm.nih.gov/pmc/articles/PMC5297920/>/articles/PMC5297920/<https://www.ncbi.nlm.nih.gov/pmc/articles/PMC5297920/> HTH Dave -- Dr David C Briggs Hohenester Lab Department of Life Sciences Imperial College London UK http://about.me/david_briggs ________________________________ From: CCP4 bulletin board <[email protected]> on behalf of Zhijie Li <[email protected]> Sent: Friday, July 28, 2017 8:59:00 PM To: [email protected] Subject: Re: [ccp4bb] Best method for carbohydrate refinement Hi Gustavo, If I understand you correctly, you are concerned about N-glycans (N-glycosylation) on your proteins. According to your description, you have 2 protein molecules in each ASU, each bearing one potential N-glycan site. Then there are only two N-glycan sites you need to build for each dataset ( I suppose you are not going to deposit everyone of them? ). In most cases we do not see much ordered part of the N-glycans, usually just one or both of the core GlcNac (NAG) residues . If this is the case then it is not much work. In very lucky cases you may see one or two rather complete N-glycans. I think insect cells produce mostly pauci-mannose N-glycans (mostly Man3-GlcNac2, less frequently Man4-GlcNac2) possibly carrying core-fucosylation on the innermost GlcNAc. Note that although mammals only have core 1,6 fucose, insects may also have an additional core 1,3 fucose, discussed and illustrated in the following papers: https://www.ncbi.nlm.nih.gov/pmc/articles/PMC3589692/ https://www.ncbi.nlm.nih.gov/pmc/articles/PMC3647355/ I am not aware of an automated procedure for building complex glycans such as the N-glycans, so take extreme care making the correct linkages if you have to build more than the two core NAGs. The Man3Gn2-Asn should be Man_alpha1,3-[Man_alpha1,6]-Man_beta1,4-GlcNac_beta1,4-GlcNAc_beta-Asn. The core fucose(s), if present, will be alpha1,6- or alpha1,3- linked to the innermost GlcNAc. Note that the core alpha1,3 fucose modification occurs only when the core alpha1,6 is already there. Zhijie On 28/07/2017 1:43 PM, Gustavo Machado Alvares De Lima wrote: > Hello everybody, > > I was looking for suggestion on the best way to identify and refine > carbohydrates in a protein. This is the scenario: > > - I have 10 molecules in the biological assembly > - I used sf9 expression system, so I have random glycolisation and types of > glycolisations. I have some hint about possible glycolisation sites > - I believe I have only 1 glysolisation per monomer > - I have 2 molecule per AU. Space group C2 > > > I would like (even if I have to write down a script for it) to track this > glycolisation, identify the most probable ones, determine the restrictions > for each and refine them. I could do it by hand on every refinement cycle (or > a couple of cycles), but for many datasets it would take ages. > > What protocol would you suggest? > > I really appreciate any help in this subject. > > Regards, > Gustavo >
