It's always a headache when it goes to generating input files for MODELLER. Here is a scenario that I already have some good Cα templates in PDB format, I just want to use MODELLER as a builder for my all-atom models. The reason I didn't use scwrl is that the templates are not necessarily in full length, i.e., to share the same length with target sequence. In this entry I will put a bunch of example files just in case I forget everything about MODELLER.
In order to model a target, you need at least one template structure; so called "Atom files" handy and you might want to put it in a central repository. You also need an alignment file to indicate how you going to align (thread) the sequence into the template structure. Here is an example where "templ" is the template structure and "target" denotes as the target sequence. Besides, '*' is a symbol of the end of the sequence available. For more information about the meaning of every field, check the manual for detail.
>P1;templ structure:templ:19::158::::: ------------------LEDFLHTLNTFNPEFVRYEEQEDVYFEVPRPKLLRIRGVHNLKK YYLTFKEILDENNEEFYEVEFEIGDFEKAVEVFKRLGFKIQATIKKKRWVYKLNGVTLEVNR VEGIGDFVDIEVISDSPEEAKEKIWEVAKMLGLK* >P1;target sequence:target:1::179::::: AHHHHHHGSEVEIKFKIKLEDFLHTLNTFNPEFVRYEEQEDVYFEVPRPKLLRIRGVHNLKK YYLTFKEILDENNEEFYEVEFEIGDFEKAVEVFKRLGFKIQATIKKKRWVYKLNGVTLEVNR VEGIGDFVDIEVISDSPEEAKEKIWEVAKMLGLKEEDVEPRLYLELINELSGRSS*You also need to use script file as a set of commands controlling MODELLER, therefore you can run MODELLER by enter the command line 'mod script.top' to model your target. For more information of this script, check out the detail. Here is the example script where for.ali is your alignment file.
INCLUDE SET OUTPUT_CONTROL = 1 0 0 1 0 SET ALNFILE = 'foo.ali' SET KNOWNS = 'tmpl' SET SEQUENCE = 'target' SET ATOM_FILES_DIRECTORY = './somepath/' SET STARTING_MODEL= 1 SET ENDING_MODEL = 1 CALL ROUTINE = 'model'By mjhsieh at July 1, 2004 02:34 AM | Monthly Archives
please build more than one model at a time. or modeller will give you horse craps in the badly aligned region... fortunately, this alignment has no gap. this will make the model better.
The reason I want to play whatcheck is that smy talked about the quality of the loop model generated by modeller. Worrying about the loop part of predicted model, Ray suggested me to use WHATIF to check the "rough"...
Read more in WHATCHECK »