I want to use model for docking. Would it be possible to model it through modeller. Or please tell me about some threading based method.
You can use threading algorithms or ab initio prediction. But according to literature only comparative modeling with >50% of sequence identity between query and template can give model of enough quality for protein-ligand docking. So, it's a sophisticated task to make a reliable model in the absence of good template. However, if structure and sequence of active site of your enzyme are conservative you still may try to find a template and perform such modeling.
Hi,
You can use threading or protein fold recognition. If its unsuccessful, then you can try abinitio prediction. Initially try fugue server for detecting closest homologous proteins with similar fold. Based on its output sorted by Z-score, model your protein using MODELLER software.
Hi,
You can give your fasta sequence as input to FUGUE server for threading or fold recognition. Then it will generate the best possible template information sorted by Z-SCORE. It will also mention whether the template information is "certain" or "likely". Once its a certain template, then download the alignment, and processed template coordinates generated by FUGUE server. Use this info to generate your target model through MODELLER.
- Badges
- Science topic
- Similar topics
- Bioinformatics
- Bioinformatics and Computational Biology