Clustering log file gmx cluster

User discussions
1

GROMACS version:
GROMACS modification: Yes/No
Here post your question

Hi I would like to do a rmsd based clustering analysis for a protein trajectory. I am using the gromos method and gmx cluster: gmx cluster -f prot.xtc -s system.tpr -cutoff 0.1 -cl

Now, the output log file that I get, i.e. cluster.log file, the cluster centroids are written:
cl. | #st rmsd | middle rmsd | cluster members
1 | 96 0.085 | 890 .073 | 10 30 60 70 80 90 100
| | | 110 120 130 140 150 160 170
| | | 180 190 200 210 220 230 240
| | | 250 260 270 280 290 300 310
| | | 320 330 340 350 360 370 380
| | | 390 400 410 420 430 440 450
| | | 460 470 480 490 500 510 520
| | | 530 540 550 560 570 590 600
| | | 610 620 630 640 650 660 670
| | | 680 690 700 710 720 730 740
| | | 750 760 770 780 790 800 810
| | | 820 830 840 850 860 870 880
| | | 890 900 910 920 930 940 950
| | | 960 970 980 990 1000
2 | 4 0.093 | 50 .086 | 0 20 40 50
3 | 1 | 580 | 580

I would like to know whether the cluster center and member index is actually in picosecond unit, i.e. cluster center of 2nd cluster is structure at t=50 ps or it is like frame index i.e. cluster center for 2nd cluster is for 50 th frame. In my MDP file I output structure in every 10000 step. So, is the time of cluster center of 2nd cluster, t =50 ps or t = 50 x 10000 x 0.001=500 ps; as far as I understand that cluster centers and members are reported as timeframes (ps) not as frame indices. If I change the unit of time using -tu ns, values in cluster.log changes accordingly, e.g. for 2nd cluster the cluster center is t=0.05

2

Hi,

In the log the structures are labelled with time (in ps). To be sure you can check what is the largest values that you find that should correspond to the last structure of your simulations.

Best regards
Alessandra

3

Hi Alessandra, Thank you for your response. I suppose by “largest value” you mean the largest value of cluster member index which should correspond to the last structure.