VMD-L Mailing List

From: sally cii (cii.sally_at_googlemail.com)
Date: Tue Jul 17 2007 - 11:58:39 CDT

Dear all,

Thank you for all your responses, I will read the grep-like utility script
and try to use pure tcl command to accomplish grep.

As I also tried to use VMD on Unix, I installed VMD on IBM machine
(successfully) (the national HPCx), and used exceed to access X-window,
however, I found it's much slower to load the trajectory on unix than that
on windows which I suspect should not be the case. Can I ask is it normal?

Thanks in advance!
Sally Cii

On 7/17/07, John Stone <johns_at_ks.uiuc.edu> wrote:
>
> Hi,
> Try changing the script to use pure Tcl commands rather than
> calling the external grep utility (which is Unix only of course).
> A little googling sometimes goes a long way -- "tcl grep" :-)
> See the "searchFiles" procedure for ideas on how to do this in pure Tcl:
> http://wiki.tcl.tk/8405
>
> Cheers,
> John Stone
> vmd_at_ks.uiuc.edu
>
> On Tue, Jul 17, 2007 at 03:09:45PM +0100, sally cii wrote:
> > Dear Nuno,
> >
> > Thank you very much for your time. I have tried the script, it gave
> error
> > message " couldn't execute "grep": no such file or directory" after
> printing
> > out "sorting pairs of contacts ...", I think this happens because I am
> > running VMD under Windows environment other than Unix, can I ask how
> should
> > I modify the script in order to run it in Windows? I looked into the tcl
> > manual, but couldn't finger out which command should I use to replace
> grep?
> > Thank you again for your help.
> >
> > Have a good day.
> >
> > Sally Cii
> >
> >
> > On 7/17/07, Nuno Loureiro Ferreira <nunolf_at_ci.uc.pt> wrote:
> > >
> > >Hi Sally
> > >
> > >From off of my head, this script should do the job.
> > >Check it, it should work the way you want.
> > >
> > >Notice that contact pairs are appended to pairsList var. If the number
> > >of pairs is huge, probably you need more memory.
> > >Or, instead of doing the -unique sorting after all contacts have been
> > >measured (all traj frames) ,
> > >you could update the uList var after each frame computation, but it
> will
> > >slowww down the proc.
> > >I'm still not very proficient on memory usage requirements, so give it
> a
> > >try.
> > >
> > >Cheers,
> > >Nuno
> > >
> > >
> > >#######################################
> > >proc num_contact {mol sel1 sel2 dist fname} {
> > > puts "started at : [exec date]"
> > > set nf [molinfo top get numframes]
> > > set f [open $fname w]
> > > set ftmp [open ${fname}.tmp w]
> > > puts "measuring contacts ..."
> > > for {set i 0} {$i < $nf} {incr i} {
> > > $sel1 frame $i
> > > $sel1 update
> > > $sel2 frame $i
> > > $sel2 update
> > > set cList [measure contacts $dist $sel1 $sel2]
> > > foreach j [lindex $cList 0] k [lindex $cList 1] {
> > > lappend pairsList "$j $k" ;# list of contact pairs, frame
> > >$i; adds up
> > > puts $ftmp "$j $k"
> > > }
> > > }
> > > close $ftmp
> > > puts "sorting pairs of contacts ..."
> > > set uList [lsort -unique $pairsList] ;# unique contact pairs
> > > # counter for each unique pair of contacts
> > > foreach item $uList {
> > > set elem1 [lindex $item 0]
> > > set elem2 [lindex $item 1]
> > > puts $f "contact $item : [lindex [exec grep "$elem1 $elem2"
> > >${fname}.tmp | wc] 0] \toccurrence(s)"
> > > }
> > > close $f
> > > file delete ${fname}.tmp
> > > puts "ended at : [exec date]"
> > >}
> > >
> > >
> > >
> > >sally cii wrote:
> > >> Dear all,
> > >>
> > >> Thank you for all your responses.
> > >>
> > >> I have $sel1 and $sel2 imported before I loaded this script. By
> > >> correcting with set list1 [lsort -increasing $lists 0], I can get the
> > >> list sorted and saved into the file $f.
> > >>
> > >> This script is small part of what I really need to do, what I really
> > >> want is a script that does:
> > >>
> > >> For all the atoms in $sel1 there is a counter, every time an atom in
> > >> $sel2 get within certain distance ($dist) to an atom in $sel1 during
> > >> $nf frames, the corresponding counter to that atom adds 1. And the
> > >> final output I want is the values of counter for every atom in $sel1.
> > >> As I don't know how to accomplish that in one script, therefore I
> > >> tried the previous script to sort the list of $sel1 and planned to
> > >> count the index afterwards. I really appreciate if anyone can shed
> > >> some light on how to do the counting in the script for me. Thank you.
> > >>
> > >> Best regards,
> > >> Sally Cii
> > >>
> > >>
> > >> On 7/14/07, *Nuno Loureiro Ferreira* <nunolf_at_ci.uc.pt
> > >> <mailto:nunolf_at_ci.uc.pt>> wrote:
> > >>
> > >> Has Axel said, you should import those sel1 sel2 vars.
> > >> But you say that you get the result on the console, so I deduce
> > >> that you
> > >> are feeding those vars some way to the proc.
> > >>
> > >> About the no output ... because your script reports an error
> before
> > >> reaching the "puts $f".
> > >> Specifically:
> > >>
> > >> set list1 [lsort -increasing $lists 0]
> > >>
> > >> What's the "0" doing there? ;-)
> > >> And by the way, why are you sorting a list of 2 elements? Be sure
> > >you
> > >> know what you are doing, since the results
> > >> from the measure contacts command, are indices, you cannot sort
> > >these
> > >> lists, since you the correct correspondence between the elements
> > >> of both
> > >> element lists.
> > >>
> > >> Probably you want to sort the pairs of contacts, using
> -increasing
> > >and
> > >> -index 0?
> > >> Check the following:
> > >>
> > >> vmd > set c [measure contacts 3 $sel1 $sel2]
> > >> {25 25 21 22 21 22 34 22 22 34 34 35 35 23} {36 39 36 36 39 39 39
> > >> 38 56
> > >> 38 37 36 37 39}
> > >> vmd > set a [lindex $c 0]
> > >> 25 25 21 22 21 22 34 22 22 34 34 35 35 23
> > >> vmd > set b [lindex $c 1]
> > >> 36 39 36 36 39 39 39 38 56 38 37 36 37 39
> > >> vmd > foreach i $a j $b {lappend nlist "$i $j"}
> > >> vmd > puts $nlist
> > >> {25 36} {25 39} {21 36} {22 36} {21 39} {22 39} {34 39} {22 38}
> > >> {22 56}
> > >> {34 38} {34 37} {35 36} {35 37} {23 39}
> > >> vmd > lsort -increasing -index 0 $nlist
> > >> {21 36} {21 39} {22 36} {22 39} {22 38} {22 56} {23 39} {25 36}
> > >> {25 39}
> > >> {34 39} {34 38} {34 37} {35 36} {35 37}
> > >>
> > >> Is this what you want?
> > >>
> > >> Cheers,
> > >> Nuno
> > >>
> > >>
> > >>
> > >> sally cii wrote:
> > >> > Dear all,
> > >> > I want to write a tcl script that can return the list of atom
> > >> index in
> > >> > $sel1 that within a certain distance contact with atoms in
> $sel2
> > >> > during a trajectory. Here is the trial script, after I load
> this
> > >> > script, I got the atom index printed out in tcl console
> together
> > >> with
> > >> > complain about 'bad option', but not saved in the file I
> wanted.
> > >> > Can anyone tell me how to return a list in a file? thank you
> in
> > >> advance!
> > >> >
> > >> > #####################################################
> > >> > proc num_contact {mol dist fname} {
> > >> >
> > >> > set nf [molinfo top get numframes]
> > >> >
> > >> > set f [open $fname "w"]
> > >> >
> > >> > for {set i 0} {$i < $nf} {incr i} {
> > >> >
> > >> > $sel1 frame $i
> > >> >
> > >> > $sel1 update
> > >> >
> > >> > $sel2 frame $i
> > >> >
> > >> > $sel2 update
> > >> >
> > >> > set lists [measure contacts $dist $sel1 $sel2]
> > >> >
> > >> > set list1 [lsort -increasing $lists 0]
> > >> >
> > >> > puts $f "$list1"}
> > >> >
> > >> > close $f
> > >> >
> > >> > $sel1 delete
> > >> >
> > >> > $sel2 delete
> > >> >
> > >> > }
> > >> > ####################################################
> > >> >
> > >> > Sincerely,
> > >> > Sally Cii
> > >> >
> > >>
> >
> >------------------------------------------------------------------------
> > >> >
> > >> > No virus found in this incoming message.
> > >> > Checked by AVG Free Edition.
> > >> > Version: 7.5.476 / Virus Database: 269.10.4/898 - Release Date:
> > >> 12-07-2007 16:08
> > >> >
> > >>
> > >>
> > >>
> ------------------------------------------------------------------------
> > >>
> > >> No virus found in this incoming message.
> > >> Checked by AVG Free Edition.
> > >> Version: 7.5.476 / Virus Database: 269.10.6/902 - Release Date:
> > >15-07-2007 14:21
> > >>
> > >
> > >
>
> --
> NIH Resource for Macromolecular Modeling and Bioinformatics
> Beckman Institute for Advanced Science and Technology
> University of Illinois, 405 N. Mathews Ave, Urbana, IL 61801
> Email: johns_at_ks.uiuc.edu Phone: 217-244-3349
> WWW: http://www.ks.uiuc.edu/~johns/ Fax: 217-244-6078
>