Can I develop homology model for DPP-8 and DPP-9 proteins? their ID are >gi|11095187 and >gi|90082057. Actually they are nucleotide sequences.
I am working on development of novel DPP-4 inhibitors and want to develop the homology model for DPP-8 and DPP-9 proteins. They are homologous to DPP-4 and there are number of crustal structures available for DPP-4 but non for DPP-8 and 9. The ligands should be selective towards DPP-4 other wise they show toxicity.
My aim to develop the homology model for DPP-8 and 9 and use it for predicting the selectivity of DPP-IV ligands as assay kit for DPP-8 and 9 are too costly.
Any advice?
Thanks
Hi Omprakash
Theoretically you should be able to do the homology modelling of the proteins of your interest....and it will definitely depend on the extent of similarity between your sequence and the template molecule (in your case DPP-4)...which you can use for the homology modeling....There are several programs (both online and offline) available for such studies....go throught them.....but....always keep in mind the final experimental validations for the predicted structure..
Ajay
Hello Omprakash,
With the VLIFE MDS 4.1 version at your place you can build the homology model using normal homology modeling technique also i would suggest the use of Multi Template technique.
Defiantly as pointed out by Ajay you have to validate the generated model.
Thank you Ajay,
Although there are a number of tools available for this work but I know a little about this. Which is the best tool for homology modeling. I will be using the model for docking purpose so the model should be robust, and as you said that model validation is must. Also I want to know hot to get the amino acid sequence from gen bank ID.
Hi Omprakash
You are right in mentioning that there are a lot of tools available in the public domain for homology modeling..I'll not go for one tool just based on random selection.....infact .....I'll suggest you to get hold of one based on the recent literature available where people might have used certain tool for solving the problem similar to yours.....and see what conclusions they could have drawn out of that analysis...
...and even if you spend extra time by doing homology modeling by two tools/software...it would be nice...
..Experimental validation definitely proves whatever you have projected based on the homology modeling...
As far as aminoacid sequence is concerned...for the genBank IDs you were mentioning....there should be a link to it...or else you translate in 'in silico' using any general purpose sequence analysis program ....online or offline...it is simple...
bye
Ajay
There are free source stand alone and web based tools available. You can refer to some of the bioinformatics tool directory like @ service section of www.bisa.co.in Hope this helps!
You can do homology modeling with swissmodel server which will give you model along with validation results
I suggest you to try several servers, there are many, check the CASP results for more information on their success rate.
Taking toghether the results form several CASPS, some top servers would be I-TASSER, ROBETTA, SAM-T08, PCONS. Swissmodel is somehow old, but still you can use it to try your models qualty (you do not need to produce the model there to test it there).
Check these two papers:
http://www.plosone.org/article/info%3Adoi%2F10.1371%2Fjournal.pone.0012483
http://www.cicy.mx/Sitios/Journal/pdf/2011/October/HighDefinition/IBCJ_2011_1_1_26S.pdf
The second is published in an unindexed journal that is just starting, because i want to help the journal, yet, is a good paper and can give a good advice.
I would recommend PHYRE2. For me it was much more efficient than Swissmodel.
- Badges
- Science topic
- Similar topics
- Bioinformatics
- Bioinformatics and Computational Biology