Dear Sir/Madam,
I have several questions regarding clustering in ptraj, could you mind give me some more advice?
I have done long MD for a protein with different mutations at residue 1000 (wild, mutantA, mutantB, mutantC). I want to do clustering analysis on rms of this protein for the first 999 residues, to see if the conformations sampled in wildtype is similar to mutantA, mutantB or mutantC.
By reading the manual, I found the commands to do clustering for each trajectories, such as:
trajin wildtype.gzcluster out testcluster representative pdb \average pdb averagelinkage clusters 4 rms .CA
However, when I finish the analysis for wild-type, how to can compare them with mutant A. If I understand correctly, when I do the following analysis, the four clusters in wild type is not necessarily same as those in mutant A. As a result, I still cannot find out the similarity between wildtype and mutant A, B and C.
trajin mutantA.gzcluster out testcluster representative pdb \average pdb averagelinkage clusters 4 rms .CA
===========================================
However, how can I do the analysis for four different trajectories? I tried to modify the input file as follows, but I think there are a lot of problems here. Could you mind to comments?
trajin wildtype.gztrajin mutantA.gztrajin mutantB.gztrajin mutantC.gzcluster out testcluster representative pdb \average pdb averagelinkage clusters 4 rms 1:999.CA
I can forsees this input file could not work because wildytpe and mutant AtoC are using different prmtop files, I cannot call up ptraj ../xxx.prmtop < xxx.ptraj command.
Could experts please kindly advice some workable and smarter approach to do clustering analysis on structures from four different trajectories?
Best regards,
Catherine
_______________________________________________
AMBER mailing list
AMBER.ambermd.org
http://lists.ambermd.org/mailman/listinfo/amber
Received on Tue Aug 09 2011 - 11:00:02 PDT
