[gmx-users] g_cluster failing

[Mark Abraham]

Building all-against-all matrices takes a lot of memory, and if it runs
out, then crazy things happen. I suggest you start with versions of your
trajectories sampled very sparsely so there are few tens of samples, and
get a sane result from g_cluster from that. Then scale up / find a lot of
memory / etc.

Mark


On Tue, Feb 18, 2014 at 5:03 PM, Davide Mercadante <
dmer018 at aucklanduni.ac.nz> wrote:

> Hello GMX users,
>
> I hope you could help me solving this un-understandable (at least for me)
> error.
>
> I am trying to run g_cluster on one of my collected trajectories using the
> following command
>
> g_cluster -f allnj_1.xtc -n index.ndx -o -g -dist -ev -sz -tr -ntr -clid
> -cl -cutoff 0.5 -skip 10 -s ../analysis_Pnup153_1/protein.gro
>
> However, I encounter the following error at the end of the g_cluster matrix
> calculations
>
> Select a group: 3
> Selected 3: 'C-alpha'
>
> Select group for output:
> Group     0 (         System) has 23957 elements
> Group     1 (        Protein) has   973 elements
> Group     2 (      Protein-H) has   502 elements
> Group     3 (        C-alpha) has    79 elements
> Group     4 (       Backbone) has   237 elements
> Group     5 (      MainChain) has   315 elements
> Group     6 (   MainChain+Cb) has   384 elements
> Group     7 (    MainChain+H) has   382 elements
> Group     8 (      SideChain) has   591 elements
> Group     9 (    SideChain-H) has   187 elements
> Group    10 (    Prot-Masses) has   973 elements
> Group    11 (    non-Protein) has 22984 elements
> Group    12 (          Water) has 22956 elements
> Group    13 (            SOL) has 22956 elements
> Group    14 (      non-Water) has  1001 elements
> Group    15 (            Ion) has    28 elements
> Group    16 (             NA) has    14 elements
> Group    17 (             CL) has    14 elements
> Group    18 ( Water_and_ions) has 22984 elements
> Select a group: 1
> Selected 1: 'Protein'
> Last frame      15000 time 150000.000
> Allocated 17630760 bytes for frames
> Read 1501 frames from trajectory allnj_1.xtc
> Computing 1501x1501 RMS deviation matrix
> # RMSD calculations left: 0
>
> The RMSD ranges from 0.0608911 to 2.57496 nm
> Average RMSD is 0.955617
> Number of structures for matrix 1501
> Energy of the matrix is 573.219 nm
> WARNING: rmsd minimum 0 is below lowest rmsd value 0.0608911
>
> Back Off! I just backed up rmsd-dist.xvg to ./#rmsd-dist.xvg.2#
> Linking structures *************
> Sorting and renumbering clusters
>
> Found 4 clusters
>
> Writing middle structure for each cluster to clusters.pdb
>
> Back Off! I just backed up clusters.pdb to ./#clusters.pdb.2#
> Counted 3 transitions in total, max 0 between two specific clusters
>
> Back Off! I just backed up clust-trans.xpm to ./#clust-trans.xpm.2#
>
> -------------------------------------------------------
> Program g_cluster, VERSION 4.5.5
> Source code file: /users/costesbn/build/gromacs-4.5.5/src/gmxlib/matio.c,
> line: 952
>
> Fatal error:
> hi (0.000000) <= lo (0.000000)
> For more information and tips for troubleshooting, please check the GROMACS
> website at http://www.gromacs.org/Documentation/Errors
> -------------------------------------------------------
>
> Can you please help me to understand what is going wrong?
>
> I have successfully used the same command to g_cluster other trajectories.
> Here for some reasons is not working.
>
> Thanks in advance to any helper.
>
> Cheers,
> Davide
> --
> Gromacs Users mailing list
>
> * Please search the archive at
> http://www.gromacs.org/Support/Mailing_Lists/GMX-Users_List before
> posting!
>
> * Can't post? Read http://www.gromacs.org/Support/Mailing_Lists
>
> * For (un)subscribe requests visit
> https://maillist.sys.kth.se/mailman/listinfo/gromacs.org_gmx-users or
> send a mail to gmx-users-request at gromacs.org.
>