Dear Sir/Madam,
I am learning how to do cluster based on Prof. Rosswalker's tutorial.
http://www.rosswalker.co.uk/tutorials/amber_workshop/Tutorial_eight/section6.htm
For my own study, I have converted my mdcrd files to one single binpos using ptraj. This binpos file should include all the atoms in the trajectory (0 to 100 ps), i.e. residues 1 to 60000.
Then, i want to convert them into PDB file with the followings conditions.
(1) As memory problems raised, I want to cut the number of PDB files to be analysis by 5 folds, i.e. only one out of the five snapshots were converted to a PDB file. Let's say a total of 100 snapshots were stored in the mdcrd file, I only want to choose 1, 5th, 10th, 15th.....snapshots to be converted to PDB format.
How to do modify this commands to get this results?
trajin xxx.mdcrd.gztrajout xxx.binpos binpos
(2) Out of the 60000 residues, I am only interested to study clustering analysis based on the rmsd or contact analysis for the residue 200-1000 and 3000 to 3500. Should I do it in the step where I convert file to PDB or should I do it when I in the clustering step? Could you mind to give me a hand how to do so? As I found it difficult to I am now interested to analysis the two loops in a proteins that is not directly connected. In the following command line, how can I define I am only interested to study the rmsd for residue 200-1000 and 3000-3500?
kclust -mode rmsd -centroid -cdist -heavy -lsqfit \
-radius 6 -maxerr 1 -iterate \
../clustfils > ../Centroid_6
Best regards,
Catherine
_______________________________________________
AMBER mailing list
AMBER.ambermd.org
http://lists.ambermd.org/mailman/listinfo/amber
Received on Wed Aug 10 2011 - 03:00:03 PDT
