[gmx-users] g_cluster analysis for structure refirement

[James Starlight]

Thanks!


That was very usefull. BTW have someone tried some software for
visualization of such clusters and compution of the percents of the SS
elements in each cluster (for instance by comparison of the most
representative structures from each cluster etc). Might this VMD plugin
http://physiology.med.cornell.edu/faculty/hweinstein/vmdplugins/clustering/
be usefull for such task?


James


2014-07-26 11:40 GMT+04:00 Thomas Evangelidis <tevang3 at gmail.com>:

> On 26 July 2014 10:15, James Starlight <jmsstarlight at gmail.com> wrote:
>
> > Hi Tsjerk,
> >
> >
> > thanks for suggestions! Some additional questions regarding g_cluster
> > method.
> >
> > 1- how most correctly select cutoff value for clustering of the
> > surface-exposed loops assuming big number of clusters? In literature I've
> > found cutoff ~ 1.3 A for enzymes active-site loops applicable for
> analysis
> > of the md trajectory. In my case I suppose cutoff might be higher due to
> > the Modeller loop prediction based on the SA (which gives bigger RMSD
> > variance of the DATA) as well as that fact that loops in my case are
> > surface expoced.
> >
> >
> You decide about the cutoff based on the final cluster populations.
>
>
>
> > 2- What flag in g_cluster could be used to save from each cluster of its
> > most representative (not averaged!) structure as the individual pdb?
> >
> >
> "-cl" , but without using the "-av" flag.
>
>
>
> > Thanks a lot,
> >
> > James
> >
> >
> > 2014-07-25 12:56 GMT+04:00 Tsjerk Wassenaar <tsjerkw at gmail.com>:
> >
> > > Hi James,
> > >
> > > The first part is just conformational clustering, for which you can use
> > > g_cluster. The easiest is then to collect the structures belonging to
> the
> > > different clusters in different files (which g_cluster can do), which
> you
> > > process separately to extract those properties you're interested in.
> Then
> > > you can easily plot the selected properties per cluster.
> > >
> > > Hope it helps,
> > >
> > > Tsjerk
> > >
> > >
> > > On Fri, Jul 25, 2014 at 9:53 AM, James Starlight <
> jmsstarlight at gmail.com
> > >
> > > wrote:
> > >
> > > > Dear Gromacs users!
> > > >
> > > > I'd like to use g_cluster utility to cluster a big set of models
> > produced
> > > > by Modeller as the result of the refirement of some parts of my
> > protein.
> > > In
> > > > this case all structures differs only in the conformation of 1
> longest
> > > loop
> > > > (~ 30 amino acids including 2 disulphide bridges) so I need to
> cluster
> > > all
> > > > models based on the RMSD in this region.   As the result I'd like to
> > > obtain
> > > > projection of all set of conformers onto the plane of some
> coordinates
> > > >  correspond to some selected structural criteriums (for instance
> > percent
> > > of
> > > > the occurence of secondary structure elements  in the refined (loop)
> > > > region; and/or some geometrical criteriums like distance between pair
> > of
> > > > residues, occurence of the salt-bridges in this region etc. As the
> > result
> > > > I'd like to visualize data and chose most representative structures
> > from
> > > > each cluster for further MD simulation. Could such processing be
> > > performed
> > > > by g_cluster taking into account that I have gromacs-like trr
> consisted
> > > of
> > > > my conformers ? On what cluster algoritms and parameters  should I
> pay
> > a
> > > > lot of attention?
> > > >
> > > > Many thanks for suggestion,
> > > >
> > > > James
> > > > --
> > > > 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.
> > > >
> > >
> > >
> > >
> > > --
> > > Tsjerk A. Wassenaar, Ph.D.
> > > --
> > > 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.
> > >
> > --
> > 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.
> >
>
>
>
> --
>
> ======================================================================
>
> Thomas Evangelidis
>
> PhD student
> University of Athens
> Faculty of Pharmacy
> Department of Pharmaceutical Chemistry
> Panepistimioupoli-Zografou
> 157 71 Athens
> GREECE
>
> email: tevang at pharm.uoa.gr
>
>           tevang3 at gmail.com
>
>
> website: https://sites.google.com/site/thomasevangelidishomepage/
> --
> 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.
>