Dear GROMACS community, I am trying to calculate B-factors of a sample equilibrated system (31 frames). My trajectory is in .psf and .dcd format but I have created a .tpr file for my analysis from the first frame . First I eliminated random translation and rotation of my protein by fitting every frame to the first one (I selected to fit CA atoms and write the whole system back):
trjconv -f truncated_WT.dcd -o truncated_WT_fitted.trr -s truncated_initial_WT.tpr -fit rot+trans Then I tried to calculate the average coordinates and save then in a .pdb file, either by first fitting all the frames to the first one (I selected to fit CA atoms): g_rmsf -f truncated_WT_fitted.trr -s truncated_initial_WT.tpr -ox -fit or without doing a least squares superposition (I selected to fit CA atoms - bizarre) : g_rmsf -f truncated_WT_fitted.trr -s truncated_initial_WT.tpr -ox -nofit However, in the latter case I am always prompted to select group(s) for root mean square calculation, and besides that the average coordinates written to the xaver.pdb file are incomplete. These are the first 10 lines: ==> xaver_fit_grmsf.pdb <== TITLE vmdmolecule0 REMARK THIS IS A SIMULATION BOX CRYST1 125.654 115.248 133.144 90.00 90.00 90.00 P 1 1 MODEL 1 ATOM 5 CA MET 1 98.823 64.559 100.152 1.00408.90 ATOM 24 CA PRO 2 98.801 61.706 101.694 1.00232.79 ATOM 38 CA PRO 3 96.250 62.337 103.868 1.00106.03 ATOM 50 CA ARG 4 92.760 61.521 102.890 1.00 69.93 ATOM 76 CA PRO 5 89.888 63.926 103.105 1.00 71.21 ATOM 88 CA SER 6 88.220 63.301 106.384 1.00 41.38 ==> xaver_nofit_grmsf.pdb <== TITLE vmdmolecule0 REMARK THIS IS A SIMULATION BOX CRYST1 125.654 115.248 133.144 90.00 90.00 90.00 P 1 1 MODEL 1 ATOM 5 CA MET 1 98.819 -nan 100.135 1.00407.71 ATOM 24 CA PRO 2 98.801 -nan 101.677 1.00231.97 ATOM 38 CA PRO 3 96.249 -nan 103.853 1.00105.71 ATOM 50 CA ARG 4 92.760 -nan 102.876 1.00 69.63 ATOM 76 CA PRO 5 89.886 -nan 103.092 1.00 70.84 ATOM 88 CA SER 6 88.220 -nan 106.373 1.00 41.21 The same happens no matter what group I select for fitting and or which version of GROMACS Tools I use (4.5.4, 4.0.5). I also tried to use .xtc and .pdb file formats instead of .trr and .trp without any luck. Surprisingly I could get complete coordinates from g_covar v4.5.4 without fitting, without being prompted to choose a group for the least squares fit. However, g_covar is much slower than g_rmsf and I would avoid using it for the whole trajectory if I could. For the sake of comparison these are the average coordinates I got from g_covar with and without fitting (file names are self-explanatory): ==> average_fit_gcovar.pdb <== TITLE Average structure MODEL 1 ATOM 5 CA MET 1 31.396 3.841 9.383 1.00 0.00 ATOM 24 CA PRO 2 31.375 0.988 10.924 1.00 0.00 ATOM 38 CA PRO 3 28.823 1.620 13.099 1.00 0.00 ATOM 50 CA ARG 4 25.334 0.802 12.120 1.00 0.00 ATOM 76 CA PRO 5 22.461 3.208 12.336 1.00 0.00 ATOM 88 CA SER 6 20.794 2.581 15.615 1.00 0.00 ATOM 99 CA SER 7 19.269 4.709 18.340 1.00 0.00 ATOM 110 CA GLY 8 21.111 5.337 21.583 1.00 0.00 ==> average_nofit_gcovar.pdb <== TITLE Average structure MODEL 1 ATOM 5 CA MET 1 98.815 64.568 100.152 1.00 0.00 ATOM 24 CA PRO 2 98.794 61.715 101.694 1.00 0.00 ATOM 38 CA PRO 3 96.243 62.347 103.869 1.00 0.00 ATOM 50 CA ARG 4 92.753 61.529 102.890 1.00 0.00 ATOM 76 CA PRO 5 89.880 63.935 103.105 1.00 0.00 ATOM 88 CA SER 6 88.213 63.308 106.385 1.00 0.00 ATOM 99 CA SER 7 86.688 65.436 109.109 1.00 0.00 ATOM 110 CA GLY 8 88.530 66.064 112.353 1.00 0.00 I also visualized truncated_WT_fitted.trr and average_nofit_gcovar.pdb in VMD and concluded that the average coordinates of average_nofit_gcovar.pdb are most likely correct (you can also work it out by comparing the coordinates of average_nofit_gcovar.pdb, xaver_fit_grmsf.pdb and xaver_nofit_grmsf.pdb). Any ideas what I am doing wrong with g_rmsf? thanks in advance, Thomas
-- gmx-users mailing list gmx-users@gromacs.org http://lists.gromacs.org/mailman/listinfo/gmx-users Please search the archive at http://www.gromacs.org/Support/Mailing_Lists/Search before posting! Please don't post (un)subscribe requests to the list. Use the www interface or send it to gmx-users-requ...@gromacs.org. Can't post? Read http://www.gromacs.org/Support/Mailing_Lists
