In silico modeling of arch-3: Medeller and RosettaCM model building.

<p>Here are all the files nessesary for prediction of archaerhodopsin-3 structure using Medeller or RosettaCM algorithms. </p> <p>For modeling using  Medeller  please read README_MEDELLER file for all the information. For RosettaCM:</p> <p>Please, before using these scripts adjust them for you cluster. You will need the recent version of Rosetta package installed.</p> <p>Let's take for example stucture P96787 (arch-3) as a query, 1UAZ_full.pdb as a template.</p> <p>Also, you'll need grishex.script, thread.script, hybridize.script, relaxate.script, stage1_membrane.wts, stage2_membrane.wts, stage3_rlx_membrane.wts, uuu.grishin, rosetta_cm.options, rosetta_cm.xml, relax.options, cluster.options that are located in the  "GENERAL" folder here. Let's assume they are located in the workfolder.</p> <p>Create an alignment file. For pairwise alignement use AlignMe/MP-T. Copy the alignment information in the file P96787_1UAZ.aln in the workfolder.</p> <p>Put in the workfolder attached P96787_3.frags P96787_9.frags attached here -- fragment files for Rosetta.<br> Put in the workfolder attached P96787.octopus.</p> <p>mv 1UAZ_full.pdb 1UAZ.pdb</p> <p>Open the grish.script and check: "target" and "templateA" which describe each line of the alignment file change on the names of the lines in your alignment file!<br> For example for AlignMe it will be P96787 1UAZ.<br> ./grishex.script P96787 1UAZ</p> <p>./thread.script P96787 1UAZ</p> <p><br> Run the hybridize script: ./hybridize.script P96787 1UAZ 500<br> It will take several days.</p> <p>It will create file hybridized_P96787.out -- a binary silent file of 500 structures with the rebuilt loops and inserted fragments, if<br> there were gaps. It will take several days also.</p> <p>Run the ./relax.script P96787 50<br>  <br> It will create file relaxed_P96787.out -- a binary silent file of 25000 structures -- 50 for each hybridized.</p> <p>Time to evaluate the results.</p> <p>Clustering.</p> <p>Run ./do.script P96787<br> open LISTIK<br> delete the _0001 in the end of each line.<br> Run ./resulting1.script</p> <p>In the folder PDBSS open cluster_summary.txt.<br> The first structure in the list is the best one. It is located in the folder PDBSS.</p> <p><br>  </p>