if the protein identity is less than 28% through pdb blast then I can use hhpred for template search. Is this reliable.
Hello,
Seq.id 28% is still good for homology modeling, or so-called twilight-zone (seq.id
hey thanks. could u please tell me that if the protein structure does not solved by pdb then what is the prominent way to predict the structure except modeller
Hello Heena
Higher percentage of template similarity in homology modelling results a good quality of protein structure. Generally more than 35% of template similarity is consider as good. But in your case you are getting similarity less than that so you can go for modelling with multiple template approach. Otherwise you can go for threading method of modelling.
Hope this will help.
Dear Heena,
I am not sure if it will help you or not. You can use iTASSER server. This server performs ab initio calculations for those which do not show structure homology with your template.
You should not consider only identity value in my opinion. Others i.e. subject coverage, query coverage, similarity and, even, the active site should be considered. For example, you do multiple alignment and see the conserved region which has been reported as an active site, therefore it makes sense to select that or those as templates.
actually my protein is new not have enough data so thats why its little bit tough to do this kind of work.
hi...heena...send me u r sequences... i can build homology model for u r target sequences.
- Badges
- Science topic