Dear Amber users
I simulated a protein for 1microsecond. This protein has 799 amino acids, a coenzyme (number 800) and acofactor (number 801) located in the active site of the protein. Then, Iclustered the trajectory in cpptraj using the following command, and I have gotthree clusters:
cluster hieragglo crdset all-trajectoriesepsilon 5 averagelinkage rms :1-800&!.H=,:801.C,N,O sieve 10 out cnumvtimesummary summaryfile info infofile clusterout clus clusterfmt amber repoutrepclus repfmt pdb cpopvtime population normframe
However, when I visualized the producetrajectory of each cluster in VMD, I found that the cofactor (number 801) movesin the active site and outside the active site with a very high RMSD. It looksthat the algorithm did not consider this cofactor in the RMS calculations, althoughit is included in the atom mask as shown in the cpptraj command.
Also I used the same command, but I changed the atoms maskas follow, and I have got the same situation:
clusterhieragglo crdset all-trajectories epsilon 5 averagelinkage rms :1-801&!.H=sieve 10 out cnumvtime summary summaryfile info infofile clusterout clusclusterfmt amber repout repclus repfmt pdb cpopvtime population normframe
1. Could you please advise me aboutthis situation and what is happening? The RMS of amino acids in the protein ineach cluster is good.
2. Can I re-cluster eachcluster based on the cofactor?
3. Is there any other methodthat I can use to find out what are the preferred locations of the cofactor inthe active site throughout the trajectory of 1 microsecond?
Thank you, indeed
Waleed
_______________________________________________
AMBER mailing list
AMBER.ambermd.org
http://lists.ambermd.org/mailman/listinfo/amber
Received on Fri Mar 25 2016 - 02:30:04 PDT
