[Dock-fans] Fwd: Re: Fwd: Re: Parallel install

Scott Brozell sbrozell at scripps.edu
Fri Sep 28 14:08:16 PDT 2007


Hi,

On Fri, 28 Sep 2007, Francesco Pietra wrote:

> Yes, amberize_ligand failed and that was reflected on the other *.out files.
> All three are attached.
>
> I can't foresee if Amber score  will become useful. Though, being somewhat
> familiar with Amber9, and having it, if there is any chance to fix this problem
> ...
>
> To this regard, why was Antechamber calling divcon at $AMBERHOME/bin/divcon?
> Which can't be found in DOCK. Divcon is available on my machine, though at
> $AMBERHOME/exe/divcon (where also antechamber is present).

Ahh, this is a known problem that has been fixed in Amber/antechamber
land, but not yet released.
The patch is to undefine AMBERHOME and rerun the tests.

Scott

> --- Scott Brozell <sbrozell at scripps.edu> wrote:
>
> > Hi,
> >
> > On Fri, 28 Sep 2007, Francesco Pietra wrote:
> >
> > > Yes, I badly forgot to carry out first serial compilation. Now done, which
> > > resulted in a seemingly major error. Please see below at your directions
> > how to
> > > carry that out.
> > >
> > > --- Scott Brozell <sbrozell at scripps.edu> wrote:
> > >
> > > > Hi,
> > > >
> > > > On Thu, 27 Sep 2007, Francesco Pietra wrote:
> > > >
> > > > > I decided to start from fresh source and compile all programs, again
> > > > setting
> > > > >
> > > > > MPICH_HOME=/usr/local
> > > > >
> > > > > as root of the installation of Open MPI
> > > > >
> > > > > Surely there was notable progress with respect to previous attempts.
> > This
> > > > time
> > > > > /bin contains a large number of executables though errors were reported
> > > > about
> > > > > anchor_and_grow. I donot understand if this is the single error or it
> > is
> > > > > truncated (probably so because the test was short)
> > > > >
> > > > > make test 2>&1 | tee test.out
> > > > >
> > > > > Attached is the compressed test.out. Againg, I am sincerely sorry to
> > convey
> > > > all
> > > > > these troubles due to my scarce experience with compilations.
> > > > >
> > > > > I was surprised to see antechamber.mpi and mopac.mpi. In Amber9 the
> > > > compilation
> > > > > of Antechamber and Divcon is serial.
> > > >
> > > > dock6 is the only parallel executable.
> > > > antechamber.mpi, etc is a minor glitch (inappropriate
> > > > file extension) that is exercised by unconventional building.
> > > > Here is the quick start for an example gnu installation:
> > > >     cd install; ./configure gnu; make install; make test;
> > >
> > > I removed all binaries generated from previous faulty parallel compilation.
> > > Then (after "make clean") I followed the above directions for serial
> > > compilation. The following binaries were generated in /bin:
> > >
> > > am1bcc, amberize_complex, amberize_ligand, amberize_receptor, antechamber,
> > > atomtype, bondtype, chemgrid, distmap, dock6, espgen, grid, grid_convert,
> > > grid_convrds, make_phimap, mopac, mopac.sh, nchemgrid_GB, nchemgrid_SA,
> > > parmcal, parmchk, prepare_amber.pl, prepgen, resp, respgen, showbox,
> > > showsphere, solvgrid, solvmap, sphere_selector, sphgen, teLeap, tleap.
> > >
> > > make test 2>&1 | tee test_serial.out (attached)
> > >
> > > revealed 29 instances of Possible Failure. I examined with make check the
> > dif
> > > files finding irrelevant differences in tails of floating points, and more
> > > difficult to understand problems with the number of conformations. Probably
> > one
> > > should be familiar with dock to fully grasp the message of these dif files.
> > It
> > > is much more complicated than with Amber9 dif.
> >
> >
> > These difs look innocuous.  Unfortunately, changes in the number of
> > Conformations cause a cascade of difs.  This is on the long term todo list.
> >
> > ./anchor_and_grow/flex2.dockout.dif
> > 58c58
> > <  Conformations:    13
> > ---
> > >  Conformations:    14
> >
> >
> > > The end of the test_serial.out reveals a possible major error (can't open
> > file
> > > lig.1.amber.pdb). I don't understand if this issue should be resolved
> > before
> > > parallel compilation (which, in fact, I have not carried out).
> >
> >
> > ../../../bin/prepare_amber.pl completed.
> > Processing test move_ligand
> > ../../../bin/dock6 -i move_ligand.dockin -o move_ligand.dockout
> > getpdb: can't open file lig.1.amber.pdb
> > make[3]: *** [move_ligand.dockout] Error 1
> >
> > If you plan to use Amber score then the missing lig.1.amber.pdb should
> > be investigated:  in install/test/amber_score_181l look at the
> > files *.out
> > It is likely that amberize_ligand failed.
> >
> > Scott
> >
> > > cd test
> > > make check 2>&1 | tee check_serial.out
> > >
> > > is also attached.
> > >
> > > From the mailing list I fear I am the single guy experiencing so much
> > troubles
> > > in compiling dock. I deeply apologize.
> > >
> > > Thanks
> > >
> > > francesco
> > >
> > >
> > >
> > >
> > > >     make distclean; ./configure gnu parallel;
> > > >     setenv MPICH_HOME /bla; make dock; cd test/mpi; make
> > > >
> > > >
> > > > Since you didnt build a serial dock, the serial dock tests are failing:
> > > > Processing test flex1
> > > > ../../../bin/dock6 -i flex1.dockin -o flex1.dockout
> > > > make[3]: ../../../bin/dock6: Command not found
> > > >
> > > >
> > > > Scott
> > > >
> > > > > --- Francesco Pietra <chiendarret at yahoo.com> wrote:
> > > > >
> > > > > > Date: Thu, 27 Sep 2007 02:26:48 -0700 (PDT)
> > > > > > From: Francesco Pietra <chiendarret at yahoo.com>
> > > > > > Subject: Fwd: Re: [Dock-fans] Fwd: Re:  Parallel install
> > > > > > To: Scott Brozell <sbrozell at scripps.edu>
> > > > > > CC: dock-fans <dock-fans at docking.org>
> > > > > >
> > > > > > I forgot to mention that I plan to use Antechamber from Amber9.
> > > > Summarizing
> > > > > > this and questions below:
> > > > > > (1) What to do with legacy_accessories.tar.gz?
> > > > > > (2) Delete antechamber and mopac from /src before compiling dock? Or
> > what
> > > > > > better in order to avoid installing antechamber to save spece
> > > > > > Thanks
> > > > > > francesco
> > > > > >
> > > > > >
> > > > > > --- Francesco Pietra <chiendarret at yahoo.com> wrote:
> > > > > >
> > > > > > > Date: Thu, 27 Sep 2007 00:55:57 -0700 (PDT)
> > > > > > > From: Francesco Pietra <chiendarret at yahoo.com>
> > > > > > > Subject: Re: [Dock-fans] Fwd: Re:  Parallel install
> > > > > > > To: Scott Brozell <sbrozell at scripps.edu>
> > > > > > > CC: dock-fans <dock-fans at docking.org>
> > > > > > >
> > > > > > > Please, read below. My main question before cleaning and
> > re-install, is
> > > > > > what
> > > > > > > to
> > > > > > > do with
> > > > > > >
> > > > > > >
> > > > > > > legacy_accessories.tar.gz
> > > > > > >
> > > > > > >
> > > > > > > --- Scott Brozell <sbrozell at scripps.edu> wrote:
> > > > > > >
> > > > > > > > Hi,
> > > > > > > >
> > > > > > > > Did you read the test output ?
> > > > > > >
> > > > > > > Yes, I noticed that, though I was unable to decide what to do.
> > Please
> > > > see
> > > > > > > below.
> > > > > > > >
> > > > > > > > ...
> > > > > > > > make[3]: Entering directory
> > > > `/usr/local/dock6/install/test/bump_filter'
> > > > > > > > cd ../grid_generation && make test
> > > > > > > > make[4]: Entering directory
> > > > > > `/usr/local/dock6/install/test/grid_generation'
> > > > > > > > # Construct box to enclose spheres
> > > > > > > > ../../../bin/showbox < box.in > /dev/null
> > > > > > > > /bin/sh: ../../../bin/showbox: No such file or directory
> > > > > > > >
> > > > > > > > Obviously, you should confirm that bin/showbox does not exist
> > > > > > > > and investigate its installation.
> > > > > > > >
> > > > > > > > Scott
> > > > > > >
> > > > > > >
> > > > > > > /usr/local/dock6//bin only contains two binary files, dock6
> > (generated
> > > > by
> > > > > > > serial install) and dock6.mpi (generated by parallel install).
> > Details
> > > > > > below.
> > > > > > >
> > > > > > > The download contained
> > > > > > > dock.6.1_source.tar.gz
> > > > > > > and
> > > > > > > legacy_accessories.tar.gz
> > > > > > >
> > > > > > >
> > > > > > > I did nothing with the latter tar.gz file.
> > > > > > >
> > > > > > > Decompressing dock.6.1.... into
> > > > > > >
> > > > > > > /usr/local
> > > > > > >
> > > > > > > as described in the manual, gave
> > > > > > > /usr/local/dock6
> > > > > > > containing:
> > > > > > > bin;
> > > > > > > install;
> > > > > > > parameters;
> > > > > > > src (accessories, antechamber, dock, docktools, gbsa_grids, grid,
> > > > mopac6,
> > > > > > > resp,
> > > > > > > tleap, Makefile);
> > > > > > > tutorials.
> > > > > > >
> > > > > > > As I am the only user of the computing machine, I did:
> > > > > > >
> > > > > > > chown -R francesco dock6
> > > > > > > chgrp -R francesco dock6
> > > > > > >
> > > > > > > Install g++ 4.1.2 from Debian package (g77-doc g77-3.4 gcc-3.4
> > > > lib2c0-dev)
> > > > > > > (as
> > > > > > > Open MPI was compiled with Intel, like Amber9)
> > > > > > >
> > > > > > > ./configure gnu
> > > > > > >
> > > > > > > make dock # builds only dock programs
> > > > > > > make dock # builds the accessory programs
> > > > > > >
> > > > > > > I made the test and probably overlooked the problems (see later)
> > > > > > >
> > > > > > > Then configure parallel:
> > > > > > >
> > > > > > >
> > > > > > > ./configure gnu parallel
> > > > > > >
> > > > > > > Then set
> > > > > > >
> > > > > > > MPICH_HOME=/usr/local
> > > > > > >
> > > > > > > Then
> > > > > > > make clean
> > > > > > > make dock # builds only the dock program
> > > > > > > make utils # builds only the accessory programs
> > > > > > >
> > > > > > >
> > > > > > >
> > > > > > > Then
> > > > > > > make  clean
> > > > > > > make test 2>&1 | tee test.out
> > > > > > > make check 2>&1 | tee check.out
> > > > > > >
> > > > > > > Surely there was a mistake in the serial install, where
> > > > > > >
> > > > > > > make dock # builds the accessory programs
> > > > > > > should have been
> > > > > > > make utils # builds the accessory programs
> > > > > > >
> > > > > > > I guess the best is deleting everything and start from fresh
> > > > > > > dock.6.1_source.tar.gz, first serial, then reliable make test, then
> > > > > > parallel.
> > > > > > > Though, I wait until I know what to do with the legacy file where I
> > see
> > > > > > many
> > > > > > > fortran sources.
> > > > > > >
> > > > > > > Thanks
> > > > > > > francesco
> > > > > > >
> > > > > > >
> > > > > > > >
> > > > > > > > On Wed, 26 Sep 2007, Francesco Pietra wrote:
> > > > > > > >
> > > > > > > > > Attached the tests according to the correct synthax for
> > generating
> > > > .out
> > > > > > > > >
> > > > > > > > > Thanks
> > > > > > > > > francesco
> > > > > > > > >
> > > > > > > > >
> > > > > > > > > --- Scott Brozell <sbrozell at scripps.edu> wrote:
> > > > > > > > >
> > > > > > > > > > Hi,
> > > > > > > > > >
> > > > > > > > > > On Wed, 26 Sep 2007, Francesco Pietra wrote:
> > > > > > > > > >
> > > > > > > > > > > The whole thread is below for record, though, new events
> > are
> > > > all
> > > > > > > > summarized
> > > > > > > > > > > here.
> > > > > > > > > > >
> > > > > > > > > > > There are two compilations, dock6 and dock6.mpi, in:
> > > > > > > > > > >
> > > > > > > > > > > /usr/local/dock6/bin
> > > > > > > > > > >
> > > > > > > > > > > I stick at
> > > > > > > > > > >
> > > > > > > > > > > MPICH_HOME=/usr/local
> > > > > > > > > > >
> > > > > > > > > > > Then I delete the outputs of previous "make test" and carry
> > out
> > > > new
> > > > > > > > ones as
> > > > > > > > > > > follows
> > > > > > > > > > >
> > > > > > > > > > > (1) From the directory
> > > > > > > > > > > /usr/local/dock6/install
> > > > > > > > > > >
> > > > > > > > > > > make clean
> > > > > > > > > > > make test | tee test.out
> > > > > > > > > > >
> > > > > > > > > > > (2) From the directory
> > > > > > > > > > >
> > > > > > > > > > > /usr/local/dock6/install/test
> > > > > > > > > > >
> > > > > > > > > > > make check | tee check.out
> > > > > > > > > > >
> > > > > > > > > > > The two output files are attached. Again, test.out is very
> > > > short
> > > > > > and
> > > > > > > > looks
> > > > > > > > > > very
> > > > > > > > > > > similar to previous trials. I am quite sorry to give these
> > > > > > troubles.
> > > > > > > > > >
> > > > > > > > > >
> > > > > > > > > > The tests are failing as before.  The error message is not
> > being
> > > > > > > > > > redirected.  Watch the actual full output as it is emitted or
> > > > > > > > > > use the correct syntax to pipe stderr (for bash):
> > > > > > > > > >
> > > > > > > > > > make test 2>&1 | tee test.out
> > > > > > > > > >
> > > > > > > > > > Scott
> > > > > > > > >
> > > > > > > > >
> > > > > > > >
> > > > > > >
> > > > > > >
> > > > > > >
> > > > > > >
> > > > > > >
> > > > > >
> > > > >
> > > >
> > >
> >
> ____________________________________________________________________________________
> > > > > > > Fussy? Opinionated? Impossible to please? Perfect.  Join Yahoo!'s
> > user
> > > > > > panel
> > > > > > > and lay it on us.
> > > > > > http://surveylink.yahoo.com/gmrs/yahoo_panel_invite.asp?a=7
> > > > > > >
> > > > > > >
> > > > > > >
> > > > > >
> > > > > >
> > > > > >
> > > > > >
> > > > > >
> > > > > >
> > > > >
> > > >
> > >
> >
> ____________________________________________________________________________________
> > > > > > Yahoo! oneSearch: Finally, mobile search
> > > > > > that gives answers, not web links.
> > > > > > http://mobile.yahoo.com/mobileweb/onesearch?refer=1ONXIC
> > > > > >
> > > > >
> > > > >
> > > > >
> > > > >
> > > > >
> > > > >
> > > >
> > >
> >
> ____________________________________________________________________________________
> > > > > Need a vacation? Get great deals
> > > > > to amazing places on Yahoo! Travel.
> > > > > http://travel.yahoo.com/
> > > >
> > > >
> > >
> > >
> > >
> > >
> > >
> >
> ____________________________________________________________________________________
> > > Boardwalk for $500? In 2007? Ha! Play Monopoly Here and Now (it's updated
> > for today's economy) at Yahoo! Games.
> > > http://get.games.yahoo.com/proddesc?gamekey=monopolyherenow
> >
> >
>
>
>
>
> ____________________________________________________________________________________
> Be a better Heartthrob. Get better relationship answers from someone who knows. Yahoo! Answers - Check it out.
> http://answers.yahoo.com/dir/?link=list&sid=396545433
>



More information about the Dock-fans mailing list