I am trying to model a protein structure of known sequence of unknown structure. The sequence identity of this protein is showing maximum of 54% with the template structure. When I proceed through the swiss model workspace it is showing improper alignment of the query. Is there any suggestion for this.
If it's for sequence structure alignment, you can try align your dataset with 4SALE.
http://4sale.bioapps.biozentrum.uni-wuerzburg.de/
Cheers, HongChang
Try I-TASSER, which has been ranked as the number 1 structure prediction server in the last 4 CASP contests for protein structure modeling
http://www.jove.com/video/3259/a-protocol-for-computer-based-protein-structure-function
Hello Prem,
First I would like to know the protein of your interest is homodimer or monomer. If its monomer then you can use Zhang server (I-TASSER) as mentioned by Jeffrey. I have already worked with the same and I found that it is not suitable for homodimer or homotetramer but for monomer it works absolutely fine.
Again important issue is the identity; your protein has really good identity above 50, if its lower below 40 then there will be a chances of getting improper model.
I strongly suggest you to use ClusterW online server for alignment. Sine 3 years I was working with this server and I got good results so far. I hope you will also find it more interesting. One you get an alignment file you can develop model using Modeller9.11 which is freely available.
Best of luck.
The minimum required identity for safe homology modeling depends on the size of the protein to model. Small proteins require larger identity than large proteins. There is a curve which separates regions of safe and unsafe homology modeling depending on the size of protein.
My protein of interest is related to homohexamer, so can I use swill model in alignment mode.? as I already have done with this and I am worried whether it is reliable? does any body have any suggestion.?
and if i want to create a phylogenetic tree using protein sequence of my interest with some selected organisms' of same protein. how can I proceed, I used BioNJ tool but after entry of FASTA sequences into it error is shown by it. then how can i proceed ?
@Gaurao Dhoke
There is a new I-TASSER like program, TACOS (manuscript submitted), under development for construction of homo- and heterodimers from sequence (unfortunately, higher order oligomers are not yet available).
http://zhanglab.ccmb.med.umich.edu/TACOS/
below: native-like protein complexes reconstructed by TACOS
@Jeffrey Brender
Thanks for this useful information. I will go through it.
Generally, >30 identity means the folds are similar and can be used to model your structure.
- Badges
- Science topic
- Similar topics
- Computer Science and Engineering
- Bioinformatics
- Bioinformatic Tools