From jison at hgmp.mrc.ac.uk  Tue Feb  1 08:02:13 2005
From: jison at hgmp.mrc.ac.uk (Dr J.C. Ison)
Date: Tue, 01 Feb 2005 13:02:13 +0000
Subject: [EMBOSS] pdbparse format
References: <1106772819.3230.20.camel@localhost.localdomain>
		 <20050127162440.GA25697@iastate.edu> <1107201866.3241.21.camel@localhost.localdomain>
Message-ID: <41FF7DD5.89D06BF5@hgmp.mrc.ac.uk>

Hi Peter

The output of pdbparse is documented (the pages should be moving to 
SourceForge very soon):
http://www.rfcgr.mrc.ac.uk/Software/EMBOSS/Apps/domainatrix/pdbparse.html

The format of the CO (coordinate line) *is* likely to change soon.  At the 
moment it includes both ATOM and RESIDUE-specific data.  In a future release 
these data will be separated on different line types (this is to keep a referee
of the pdbparse ms. happy).

Any changes will be documented and code for reading and writing the new 
format will updated in AJAX (if there's a Java equivalent of "sscanf" I'd 
use that rather than rely on column positions.)

You understand the format correctly.  The 8.3 does look a bit odd but its 
fine (would have been neater to use %-3S%9.3f instead of %-4S%8.3f but there 
you go ... :)

Plus - if you have any suggestions / requirements for pdbparse, please let me 
know.

Cheers

Jon




Thanks for the interest.

Jon

Peter Robinson wrote:
> 
> Hi list,
> 
> I have been looking into the new pdbparse program and am writing a
> routine to parse the output of that program into a Java program I am
> writing to do some datamining on PDB files. For the moment at least I am
> writing a parse routine based on column positions rather than on regular
> expressions or some other approach.
> I would appreciate some help in understanding the output format. I have
> been looking at the function ajPdbWriteAll.
> 
> ajFmtPrintF(outf, "%-2c%6S    %-4S%8.3f%9.3f%9.3f<snip>,
>                                     tmp->Id1,
>                                     tmp->Id3,
>                                     tmp->Atm,
>                                     tmp->X,
>                                     tmp->Y,
>                                     tmp->Z,
>                                     <snip>
> 
> 
> this should print out the
> "%-2c":  1char AA (left-justified, 2 spaces)
> "%6S    ": 3charAA (right-justified, 6 spaces, followed by 4 spaces)
> "-4S": AtomType (e.g., CA) (left-justified, 4 spaces)
> "%8.3" x coord, 8 spaces with 3 precision
> "%9.3" y coord  9 spaces with 3 precision
> "%9.3" z coord  9 spaces with 3 precision
> 
> ==> If I am reading this correctly, why does the x coord have one less
> space than y and z?
> ==> Can one consider this output format fixed, or will it be subject to
> changes?
> ==> Excuse what may seem like a trivial question, but I don't want my
> program to break down "at an embarassing moment".
> 
> Thanks,
> Peter
> 
> --
> Peter N. Robinson
> peter.robinson at t-online.de
> peter.robinson at charite.de
> http://www.charite.de/ch/medgen/robinson/


-- 
Jon C. Ison, PhD
Proteomics Applications Group
MRC Rosalind Franklin Centre for Genomics Research
Wellcome Trust Genome Campus, Hinxton, Cambridge, CB10 1SB, UK
Tel: +44 1223 494500  Fax: +44 1223 494512
E-mail: jison at rfcgr.mrc.ac.uk  Web: http://www.rfcgr.mrc.ac.uk


From scott at cs.wits.ac.za  Thu Feb  3 11:15:43 2005
From: scott at cs.wits.ac.za (Scott Hazelhurst)
Date: Thu,  3 Feb 2005 18:15:43 +0200 (SAST)
Subject: [EMBOSS] acd file problem
Message-ID: <20050203161543.3E9A61416ED@midi.cs.wits.ac.za>



I have had problems getting a number of the embassy-type applications
to run, either through jemboss or from the command line. I have
tracked the problem down to:


Error: File /usr/local/emboss/share/EMBOSS/acd/ednaml.acd line 1: (ednaml) Attribute 'embassy' unknown

It complains about this entry in the acd file

application: ednaml [
  documentation: "Estimates phylogenies from nucleic acid
                  sequence Maximum Likelihood"
  groups: "Phylogeny:Molecular sequence"
  gui: "yes"
  batch: "yes"
  cpu: "high"
  embassy: "phylip"
]


If I delete the 'embassy: "phylip"' line, it seems to work.

What am I doing wrong?

Scott



From ableasby at hgmp.mrc.ac.uk  Thu Feb  3 11:27:28 2005
From: ableasby at hgmp.mrc.ac.uk (Alan Bleasby)
Date: Thu, 3 Feb 2005 16:27:28 GMT
Subject: [EMBOSS] acd file problem
Message-ID: <200502031627.j13GRS6d006396@bromine.hgmp.mrc.ac.uk>

>If I delete the 'embassy: "phylip"' line, it seems to work.
>What am I doing wrong?

This problem was caused by some unavoidable updates to
the embassy applications between releases. The fix
is to recompile emboss using the 'ajacd.c' file
in the:

  ftp://ftp.hgmp.mrc.ac.uk/pub/EMBOSS/patchfiles/

directory.

HTH

Alan


From pmr at ebi.ac.uk  Fri Feb  4 13:16:41 2005
From: pmr at ebi.ac.uk (Peter Rice)
Date: Fri, 04 Feb 2005 18:16:41 +0000
Subject: [EMBOSS] acd file problem
In-Reply-To: <200502031627.j13GRS6d006396@bromine.hgmp.mrc.ac.uk>
References: <200502031627.j13GRS6d006396@bromine.hgmp.mrc.ac.uk>
Message-ID: <4203BC09.7050106@ebi.ac.uk>

Alan Bleasby wrote:

>>If I delete the 'embassy: "phylip"' line, it seems to work.
>>What am I doing wrong?
> 
> 
> This problem was caused by some unavoidable updates to
> the embassy applications between releases. The fix
> is to recompile emboss using the 'ajacd.c' file
> in the:
> 
>   ftp://ftp.hgmp.mrc.ac.uk/pub/EMBOSS/patchfiles/
> 
> directory.

The embassy attribute was added so that "wossname" nkows how to find EMBASSY 
and non-EMBASSY programs. It has no other effect on the running of programs, 
so deleting it is a safe alternative.

regards,

Peter Rice



From ableasby at hgmp.mrc.ac.uk  Sat Feb 19 19:04:19 2005
From: ableasby at hgmp.mrc.ac.uk (Alan Bleasby)
Date: Sun, 20 Feb 2005 00:04:19 GMT
Subject: [EMBOSS] EMBOSS 2.10.0 released
Message-ID: <200502200004.j1K04J2l014579@bromine.hgmp.mrc.ac.uk>

EMBOSS 2.10.0 can be downloaded from the following URL:

   ftp://ftp.rfcgr.mrc.ac.uk/pub/EMBOSS/

or via the EMBOSS homepage:

   http://emboss.sourceforge.net

This release is made available to resolve issues caused by Java 1.5.0,
although it also corrects alignment numbering for some alignment
formats. The feature table handling code has also been speeded up.

Jemboss now features a beta release of Jemboss interactive
graphical output. You can select either the original PNG format or the new
graphics - the default is currently set to PNG.

For all other changes see the ChangeLog file.

Alan


From jerome at ibt.unam.mx  Thu Feb 24 18:44:49 2005
From: jerome at ibt.unam.mx (Jerome)
Date: Thu, 24 Feb 2005 17:44:49 -0600
Subject: [EMBOSS] Jemboss with openPBS?
Message-ID: <421E66F1.9000406@ibt.unam.mx>

Hi,
I'm trying to use jemboss with the scheduling server pbs. As it's 
described in the documentation, i've put this piece of code in the 
JembossAuthServer.java file:

// COMMENT THIS LINE TO USE QUEUEING SOFTWARE
/*
       boolean lforkB = aj.forkBatch(userName,passwd,environ,
                                     embossCommand,project);
*/
// UNCOMMENT ONE OF THESE LINE TO USE QUEUEING SOFTWARE
//    runAsBatch(aj,userName,passwd,project,quoteMe(embossCommand));
//    runAsGNQSBatch(aj,userName,passwd,project,quoteMe(embossCommand));
     runAsPBSBatch(aj,userName,passwd,project,quoteMe(embossCommand));

and in the runAsPBSBatch method:
String batchCommand = "/usr/local/Torque/bin/qsub "+scriptFile;

now the fact is that it don't send anything to the server. The script 
file is good writed, and i can read in the jemboss.log, this type of 
entries:

mode=batch  Thu Feb 24 17_34_02 CST 2005 cutseq -sequence cutseq0  -from 
1 -to 10 -auto
Total Memory : 13893632 bytes, Free Memory : 5362592 bytes

That seem that i don't make the batchCommand, this line :

lfork = aj.forkEmboss(userName,passwd,environ, 
          				batchCommand,project);

  I don't know really where to find the source of this problem.. Does 
someone achieve working with Pbs system ?
Thank's a lot for this program..


-- 
-- J?r?me
- Maraud, faquin, butor de pied plat ridicule !
- Ah ?.. Et moi, Cyrano Savinien-Hercule
   De Bergerac.
	(Edmond Rostand, Cyrano de Bergerac)



From jison at hgmp.mrc.ac.uk  Tue Feb  1 13:02:13 2005
From: jison at hgmp.mrc.ac.uk (Dr J.C. Ison)
Date: Tue, 01 Feb 2005 13:02:13 +0000
Subject: [EMBOSS] pdbparse format
References: <1106772819.3230.20.camel@localhost.localdomain>
		 <20050127162440.GA25697@iastate.edu> <1107201866.3241.21.camel@localhost.localdomain>
Message-ID: <41FF7DD5.89D06BF5@hgmp.mrc.ac.uk>

Hi Peter

The output of pdbparse is documented (the pages should be moving to 
SourceForge very soon):
http://www.rfcgr.mrc.ac.uk/Software/EMBOSS/Apps/domainatrix/pdbparse.html

The format of the CO (coordinate line) *is* likely to change soon.  At the 
moment it includes both ATOM and RESIDUE-specific data.  In a future release 
these data will be separated on different line types (this is to keep a referee
of the pdbparse ms. happy).

Any changes will be documented and code for reading and writing the new 
format will updated in AJAX (if there's a Java equivalent of "sscanf" I'd 
use that rather than rely on column positions.)

You understand the format correctly.  The 8.3 does look a bit odd but its 
fine (would have been neater to use %-3S%9.3f instead of %-4S%8.3f but there 
you go ... :)

Plus - if you have any suggestions / requirements for pdbparse, please let me 
know.

Cheers

Jon




Thanks for the interest.

Jon

Peter Robinson wrote:
> 
> Hi list,
> 
> I have been looking into the new pdbparse program and am writing a
> routine to parse the output of that program into a Java program I am
> writing to do some datamining on PDB files. For the moment at least I am
> writing a parse routine based on column positions rather than on regular
> expressions or some other approach.
> I would appreciate some help in understanding the output format. I have
> been looking at the function ajPdbWriteAll.
> 
> ajFmtPrintF(outf, "%-2c%6S    %-4S%8.3f%9.3f%9.3f<snip>,
>                                     tmp->Id1,
>                                     tmp->Id3,
>                                     tmp->Atm,
>                                     tmp->X,
>                                     tmp->Y,
>                                     tmp->Z,
>                                     <snip>
> 
> 
> this should print out the
> "%-2c":  1char AA (left-justified, 2 spaces)
> "%6S    ": 3charAA (right-justified, 6 spaces, followed by 4 spaces)
> "-4S": AtomType (e.g., CA) (left-justified, 4 spaces)
> "%8.3" x coord, 8 spaces with 3 precision
> "%9.3" y coord  9 spaces with 3 precision
> "%9.3" z coord  9 spaces with 3 precision
> 
> ==> If I am reading this correctly, why does the x coord have one less
> space than y and z?
> ==> Can one consider this output format fixed, or will it be subject to
> changes?
> ==> Excuse what may seem like a trivial question, but I don't want my
> program to break down "at an embarassing moment".
> 
> Thanks,
> Peter
> 
> --
> Peter N. Robinson
> peter.robinson at t-online.de
> peter.robinson at charite.de
> http://www.charite.de/ch/medgen/robinson/


-- 
Jon C. Ison, PhD
Proteomics Applications Group
MRC Rosalind Franklin Centre for Genomics Research
Wellcome Trust Genome Campus, Hinxton, Cambridge, CB10 1SB, UK
Tel: +44 1223 494500  Fax: +44 1223 494512
E-mail: jison at rfcgr.mrc.ac.uk  Web: http://www.rfcgr.mrc.ac.uk


From scott at cs.wits.ac.za  Thu Feb  3 16:15:43 2005
From: scott at cs.wits.ac.za (Scott Hazelhurst)
Date: Thu,  3 Feb 2005 18:15:43 +0200 (SAST)
Subject: [EMBOSS] acd file problem
Message-ID: <20050203161543.3E9A61416ED@midi.cs.wits.ac.za>



I have had problems getting a number of the embassy-type applications
to run, either through jemboss or from the command line. I have
tracked the problem down to:


Error: File /usr/local/emboss/share/EMBOSS/acd/ednaml.acd line 1: (ednaml) Attribute 'embassy' unknown

It complains about this entry in the acd file

application: ednaml [
  documentation: "Estimates phylogenies from nucleic acid
                  sequence Maximum Likelihood"
  groups: "Phylogeny:Molecular sequence"
  gui: "yes"
  batch: "yes"
  cpu: "high"
  embassy: "phylip"
]


If I delete the 'embassy: "phylip"' line, it seems to work.

What am I doing wrong?

Scott



From ableasby at hgmp.mrc.ac.uk  Thu Feb  3 16:27:28 2005
From: ableasby at hgmp.mrc.ac.uk (Alan Bleasby)
Date: Thu, 3 Feb 2005 16:27:28 GMT
Subject: [EMBOSS] acd file problem
Message-ID: <200502031627.j13GRS6d006396@bromine.hgmp.mrc.ac.uk>

>If I delete the 'embassy: "phylip"' line, it seems to work.
>What am I doing wrong?

This problem was caused by some unavoidable updates to
the embassy applications between releases. The fix
is to recompile emboss using the 'ajacd.c' file
in the:

  ftp://ftp.hgmp.mrc.ac.uk/pub/EMBOSS/patchfiles/

directory.

HTH

Alan


From pmr at ebi.ac.uk  Fri Feb  4 18:16:41 2005
From: pmr at ebi.ac.uk (Peter Rice)
Date: Fri, 04 Feb 2005 18:16:41 +0000
Subject: [EMBOSS] acd file problem
In-Reply-To: <200502031627.j13GRS6d006396@bromine.hgmp.mrc.ac.uk>
References: <200502031627.j13GRS6d006396@bromine.hgmp.mrc.ac.uk>
Message-ID: <4203BC09.7050106@ebi.ac.uk>

Alan Bleasby wrote:

>>If I delete the 'embassy: "phylip"' line, it seems to work.
>>What am I doing wrong?
> 
> 
> This problem was caused by some unavoidable updates to
> the embassy applications between releases. The fix
> is to recompile emboss using the 'ajacd.c' file
> in the:
> 
>   ftp://ftp.hgmp.mrc.ac.uk/pub/EMBOSS/patchfiles/
> 
> directory.

The embassy attribute was added so that "wossname" nkows how to find EMBASSY 
and non-EMBASSY programs. It has no other effect on the running of programs, 
so deleting it is a safe alternative.

regards,

Peter Rice



From ableasby at hgmp.mrc.ac.uk  Sun Feb 20 00:04:19 2005
From: ableasby at hgmp.mrc.ac.uk (Alan Bleasby)
Date: Sun, 20 Feb 2005 00:04:19 GMT
Subject: [EMBOSS] EMBOSS 2.10.0 released
Message-ID: <200502200004.j1K04J2l014579@bromine.hgmp.mrc.ac.uk>

EMBOSS 2.10.0 can be downloaded from the following URL:

   ftp://ftp.rfcgr.mrc.ac.uk/pub/EMBOSS/

or via the EMBOSS homepage:

   http://emboss.sourceforge.net

This release is made available to resolve issues caused by Java 1.5.0,
although it also corrects alignment numbering for some alignment
formats. The feature table handling code has also been speeded up.

Jemboss now features a beta release of Jemboss interactive
graphical output. You can select either the original PNG format or the new
graphics - the default is currently set to PNG.

For all other changes see the ChangeLog file.

Alan


From jerome at ibt.unam.mx  Thu Feb 24 23:44:49 2005
From: jerome at ibt.unam.mx (Jerome)
Date: Thu, 24 Feb 2005 17:44:49 -0600
Subject: [EMBOSS] Jemboss with openPBS?
Message-ID: <421E66F1.9000406@ibt.unam.mx>

Hi,
I'm trying to use jemboss with the scheduling server pbs. As it's 
described in the documentation, i've put this piece of code in the 
JembossAuthServer.java file:

// COMMENT THIS LINE TO USE QUEUEING SOFTWARE
/*
       boolean lforkB = aj.forkBatch(userName,passwd,environ,
                                     embossCommand,project);
*/
// UNCOMMENT ONE OF THESE LINE TO USE QUEUEING SOFTWARE
//    runAsBatch(aj,userName,passwd,project,quoteMe(embossCommand));
//    runAsGNQSBatch(aj,userName,passwd,project,quoteMe(embossCommand));
     runAsPBSBatch(aj,userName,passwd,project,quoteMe(embossCommand));

and in the runAsPBSBatch method:
String batchCommand = "/usr/local/Torque/bin/qsub "+scriptFile;

now the fact is that it don't send anything to the server. The script 
file is good writed, and i can read in the jemboss.log, this type of 
entries:

mode=batch  Thu Feb 24 17_34_02 CST 2005 cutseq -sequence cutseq0  -from 
1 -to 10 -auto
Total Memory : 13893632 bytes, Free Memory : 5362592 bytes

That seem that i don't make the batchCommand, this line :

lfork = aj.forkEmboss(userName,passwd,environ, 
          				batchCommand,project);

  I don't know really where to find the source of this problem.. Does 
someone achieve working with Pbs system ?
Thank's a lot for this program..


-- 
-- J?r?me
- Maraud, faquin, butor de pied plat ridicule !
- Ah ?.. Et moi, Cyrano Savinien-Hercule
   De Bergerac.
	(Edmond Rostand, Cyrano de Bergerac)