Previous Thread
Next Thread
Print Thread
Page 1 of 2 1 2
c36a4 topology file changed
#23508 02/25/10 04:36 PM
Joined: Aug 2009
Posts: 132
L
lqz Offline OP
Forum Member
OP Offline
Forum Member
L
Joined: Aug 2009
Posts: 132
Dear administrators:

My group downloaded the newest charmm edition recently and after compiling it, I tried to test it using the proteins I used. I could use the old topology files without any problem but it could not work using the new topology files. So I am trying to combine the old file and the new file. For a lot of atom types, the new file did not define. So what should I do to combine them? Should I just copy all the extra part from the old top_all22_prot.inp to the new file, or only copy the part my proteins need? If I copied all the extra part from the old file to the new top file, will it cause problem on the par file? Thanks a lot.




Last edited by lqz; 02/25/10 04:37 PM.
Re: c36a4 topology file changed
lqz #23509 02/25/10 05:03 PM
Joined: Sep 2003
Posts: 4,797
Likes: 2
Forum Member
Online Content
Forum Member
Joined: Sep 2003
Posts: 4,797
Likes: 2
Your post is not very easy to understand; what does "For a lot of atom types, the new file did not define" mean? It is often better to show actual input/output instead of trying to describe what is going on.

Topology and paramter files come in pairs which should be used together. Furthermore a PSF generated with a given RTF/PARAM file pair should always be used with this same RTF/PARAM file pair.

Your two main options are (very briefly):
1. Keep using your old RTF/PARAM/PSF files. Best choice if you want to continue an ongoing project.
or
2. Regenerate the PSF with the new RTF/PARAM files and use this combination instead. Best choice for new projects. I have no idea how old your old version is, but the organization of the toppar directory may be different in the new version. Look in the toppar/stream directory, and you may find the missing bits and pieces there.


Lennart Nilsson
Karolinska Institutet
Stockholm, Sweden
Re: c36a4 topology file changed
lennart #23511 02/25/10 06:03 PM
Joined: Aug 2009
Posts: 132
L
lqz Offline OP
Forum Member
OP Offline
Forum Member
L
Joined: Aug 2009
Posts: 132
Thanks a lot for your information. I got following three parts of warnings on the screen:

1)
CHARMM> READ PARAM UNIT 2 CARD

PARAMETER FILE BEING READ FROM UNIT 2
TITLE> *>>>>> CHARMM22 ALL-HYDROGEN PARAMETER FILE FOR PROTEINS <<<<<<<<
TITLE> *>>>>> INCLUDES PHI, PSI CROSS TERM MAP (CMAP) CORRECTION <<<<<<<
TITLE> *>>>>>>>>>>>>>>>>>>>>>> DECEMBER, 2003 <<<<<<<<<<<<<<<<<<<<<<<<<<
TITLE> * ALL COMMENTS TO ADM JR. VIA THE CHARMM WEB SITE: www.CHARMM.ORG
TITLE> * PARAMETER SET DISCUSSION FORUM
TITLE> *
PARRDR> WARNING: ATOMS IN HBOND CAP 0.00 -0.07000 1.99240 DONT EXIST
PARRDR> WARNING: ATOMS IN HBOND FA 0.00 -0.12000 1.70000 DONT EXIST
PARRDR> WARNING: ATOMS IN HBOND CN 0.00 -0.20000 1.75000 DONT EXIST
PARRDR> WARNING: ATOMS IN HBOND NC 0.00 -0.60000 1.85000 DONT EXIST
PARRDR> WARNING: ATOMS IN HBOND OCA 0.00 -0.12000 1.70000 DONT EXIST

Those parts are the new parts added into the new par file. I do not know why they were not defined.

2)

**** Warning **** The following extraneous characters
were found while command processing in PARRDR
0.000000 -0.120000 1.400000

I really do not know why that could happen.

3)
PARRDR> WARNING: ATOMS IN HBOND COA 0.00 -0.11000 2.00000 DONT EXIST
PARRDR> WARNING: ATOMS IN HBOND CF1 0.00 -0.06000 1.90000 DONT EXIST
PARRDR> WARNING: ATOMS IN HBOND CF2 0.00 -0.04200 2.05000 DONT EXIST
PARRDR> WARNING: ATOMS IN HBOND CF3 0.00 -0.02000 2.30000 DONT EXIST
PARRDR> WARNING: ATOMS IN HBOND HF1 0.00 -0.02800 1.32000 DONT EXIST
PARRDR> WARNING: ATOMS IN HBOND HF2 0.00 -0.03000 1.30000 DONT EXIST
PARRDR> WARNING: ATOMS IN HBOND F1 0.0 -0.13500 1.63000 DONT EXIST
PARRDR> WARNING: ATOMS IN HBOND F2 0.0 -0.10500 1.63000 DONT EXIST
PARRDR> WARNING: ATOMS IN HBOND F3 0.0 -0.09700 1.60000 DONT EXIST
PARRDR> WARNING: ATOMS IN HBOND C3 0.00 -0.02000 2.27500 DONT EXIST
PARRDR> WARNING: ATOMS IN HBOND CC1A 0.0 -0.06800 2.09000 DONT EXIST
PARRDR> WARNING: ATOMS IN HBOND CC1B 0.0 -0.06800 2.09000 DONT EXIST
PARRDR> WARNING: ATOMS IN HBOND CC2 0.0 -0.06400 2.08000 DONT EXIST
PARRDR> WARNING: ATOMS IN HBOND NS1 0.00 -0.20000 1.85000 DONT EXIST
PARRDR> WARNING: ATOMS IN HBOND NS2 0.00 -0.20000 1.85000 DONT EXIST

***** WARNING ***** PARRDR no nonbond parameters for atom type: 16 HF1
NO nonbond interactions will be computed for this atom type.
***** WARNING ***** PARRDR no nonbond parameters for atom type: 17 HF2
NO nonbond interactions will be computed for this atom type.
***** WARNING ***** PARRDR no nonbond parameters for atom type: 46 CN
NO nonbond interactions will be computed for this atom type.
***** WARNING ***** PARRDR no nonbond parameters for atom type: 47 CAP
NO nonbond interactions will be computed for this atom type.
***** WARNING ***** PARRDR no nonbond parameters for atom type: 48 COA
NO nonbond interactions will be computed for this atom type.
***** WARNING ***** PARRDR no nonbond parameters for atom type: 49 C3
NO nonbond interactions will be computed for this atom type.
***** WARNING ***** PARRDR no nonbond parameters for atom type: 61 NC
NO nonbond interactions will be computed for this atom type.
***** WARNING ***** PARRDR no nonbond parameters for atom type: 78 OCA
NO nonbond interactions will be computed for this atom type.
***** WARNING ***** PARRDR no nonbond parameters for atom type: 87 CF1
NO nonbond interactions will be computed for this atom type.
***** WARNING ***** PARRDR no nonbond parameters for atom type: 88 CF2
NO nonbond interactions will be computed for this atom type.
***** WARNING ***** PARRDR no nonbond parameters for atom type: 89 CF3
NO nonbond interactions will be computed for this atom type.
***** WARNING ***** PARRDR no nonbond parameters for atom type: 92 FA
NO nonbond interactions will be computed for this atom type.
***** WARNING ***** PARRDR no nonbond parameters for atom type: 93 F1
NO nonbond interactions will be computed for this atom type.
***** WARNING ***** PARRDR no nonbond parameters for atom type: 94 F2
NO nonbond interactions will be computed for this atom type.
***** WARNING ***** PARRDR no nonbond parameters for atom type: 95 F3
NO nonbond interactions will be computed for this atom type.
***** WARNING ***** PARRDR no nonbond parameters for atom type: 112 CC1A
NO nonbond interactions will be computed for this atom type.
***** WARNING ***** PARRDR no nonbond parameters for atom type: 113 CC1B
NO nonbond interactions will be computed for this atom type.
***** WARNING ***** PARRDR no nonbond parameters for atom type: 114 CC2
NO nonbond interactions will be computed for this atom type.
***** WARNING ***** PARRDR no nonbond parameters for atom type: 120 NS1
NO nonbond interactions will be computed for this atom type.
***** WARNING ***** PARRDR no nonbond parameters for atom type: 121 NS2
NO nonbond interactions will be computed for this atom type.
PARMIO> NONBOND, HBOND lists and IMAGE atoms cleared.


In fact, I checked the top file, a lot of atom types are defined there. So what is wrong?

Re: c36a4 topology file changed
lqz #23512 02/25/10 06:13 PM
Joined: Sep 2003
Posts: 4,797
Likes: 2
Forum Member
Online Content
Forum Member
Joined: Sep 2003
Posts: 4,797
Likes: 2
This behavior makes little sense to me. Which files are you reading? Did you read first the RTF then the corresponding parameter file? Have you modified the files in any way?
Create a short, but complete, input file that exhibits the problem you have. Post this file together with relevant (in this case ALL) output.


Lennart Nilsson
Karolinska Institutet
Stockholm, Sweden
Re: c36a4 topology file changed
lennart #23517 02/25/10 07:51 PM
Joined: Aug 2009
Posts: 132
L
lqz Offline OP
Forum Member
OP Offline
Forum Member
L
Joined: Aug 2009
Posts: 132
In fact, I just tried to let charmm read in the new top file and par file at the beginning of the job. It did not start reading the protein structures yet. I am building new system using protein initial structures. So I do not have RTF file yet.

When reading the topology files, the warning messages above were shown, although no such kind of warnings were shown when reading the old topology files.

Re: c36a4 topology file changed
lqz #23519 02/25/10 07:59 PM
Joined: Sep 2003
Posts: 4,797
Likes: 2
Forum Member
Online Content
Forum Member
Joined: Sep 2003
Posts: 4,797
Likes: 2
You do not create the RTF, it is part of the forcefield (eg, top_all22_prot.inp).

Sorry, cannot help you further without more information.

Here is what I get when reading the protein RTF/PARAM files in the c35b3 release (your errors are NOT present):
CHARMM> read rtf card name /applic/charmm/c35b3/toppar/top_all22_prot.inp
VOPEN> Attempting to open::/applic/charmm/c35b3/toppar/top_all22_prot.inp::
MAINIO> Residue topology file being read from unit 90.
TITLE> *>>>>>>>>CHARMM22 ALL-HYDROGEN TOPOLOGY FILE FOR PROTEINS <<<<<<
TITLE> *>>>>> INCLUDES PHI, PSI CROSS TERM MAP (CMAP) CORRECTION <<<<<<<
TITLE> *>>>>>>>>>>>>>>>>>>>>>> DECEMBER, 2003 <<<<<<<<<<<<<<<<<<<<<<<<<<
TITLE> * ALL COMMENTS TO ADM JR. VIA THE CHARMM WEB SITE: www.CHARMM.ORG
TITLE> * PARAMETER SET DISCUSSION FORUM
TITLE> *
VCLOSE: Closing unit 90 with status "KEEP"

CHARMM> read param card name /applic/charmm/c35b3/toppar/par_all22_prot.inp
VOPEN> Attempting to open::/applic/charmm/c35b3/toppar/par_all22_prot.inp::

PARAMETER FILE BEING READ FROM UNIT 90
TITLE> *>>>>> CHARMM22 ALL-HYDROGEN PARAMETER FILE FOR PROTEINS <<<<<<<<
TITLE> *>>>>> INCLUDES PHI, PSI CROSS TERM MAP (CMAP) CORRECTION <<<<<<<
TITLE> *>>>>>>>>>>>>>>>>>>>>>> DECEMBER, 2003 <<<<<<<<<<<<<<<<<<<<<<<<<<
TITLE> * ALL COMMENTS TO ADM JR. VIA THE CHARMM WEB SITE: www.CHARMM.ORG
TITLE> * PARAMETER SET DISCUSSION FORUM
TITLE> *
PARMIO> NONBOND, HBOND lists and IMAGE atoms cleared.
VCLOSE: Closing unit 90 with status "KEEP"

$$$$$$ New timer profile $$$$$


Lennart Nilsson
Karolinska Institutet
Stockholm, Sweden
Re: c36a4 topology file changed
lennart #23520 02/25/10 08:06 PM
Joined: Aug 2009
Posts: 132
L
lqz Offline OP
Forum Member
OP Offline
Forum Member
L
Joined: Aug 2009
Posts: 132
So sorry for my former mistake. The following is the input file for the charmm:

* A CHARMM input script to add water to the system
* April 27 1994, adpated for HEWL by M.Buck 28/3/95 & 6/2/95
*

BOMLev -5

! Define parameters
! -----------------

set n chtr ! codename
set 6 temp/ ! data directory
set 9 out/ ! output directory

! System management
! -----------------
OPEN READ UNIT 2 CARD NAME ./top_all22_prot.inp
READ RTF UNIT 2 CARD
CLOSE UNIT 2

!
!Read nucleo-acid parameters
!
OPEN READ UNIT 2 CARD NAME ./par_all22_prot.inp
READ PARAM UNIT 2 CARD
CLOSE UNIT 2


The warnings are shown based on the scripts above.

Re: c36a4 topology file changed
lqz #23523 02/25/10 08:32 PM
Joined: Sep 2003
Posts: 4,797
Likes: 2
Forum Member
Online Content
Forum Member
Joined: Sep 2003
Posts: 4,797
Likes: 2
Very strange indeed.
Are top_all22_prot.inp and par_all22_prot.inp the original files from the c35b3 version? How was CHARMM installed? Do the testcases run? What platform (hardware, OS, compiler) are you using?

DO NOT EVER USE BOMLEV -5!!!!!


Lennart Nilsson
Karolinska Institutet
Stockholm, Sweden
Re: c36a4 topology file changed
lennart #23525 02/25/10 09:03 PM
Joined: Aug 2009
Posts: 132
L
lqz Offline OP
Forum Member
OP Offline
Forum Member
L
Joined: Aug 2009
Posts: 132
I do not know if those top and par files are the same as c35b3, but they are slightly different from that from c36a2. The warning part mostly came from the extra part of c36a4.

The charmm was installed well, as least for the serial edition. I tested it in the test/ directory accompanied in the c36a4 package. They could work and abort normally.

I used intel compilor to compile it.

Re: c36a4 topology file changed
lqz #23526 02/25/10 09:17 PM
Joined: Sep 2003
Posts: 4,797
Likes: 2
Forum Member
Online Content
Forum Member
Joined: Sep 2003
Posts: 4,797
Likes: 2
Perhaps the rtf/param files you are playing with are corrupt? Go on with your project and do not bother with c36a4 for the moment. You do realize that you have not yet told us which rtf/param files you are using, or how charmm was installed on what platform.


Lennart Nilsson
Karolinska Institutet
Stockholm, Sweden
Re: c36a4 topology file changed
lennart #23529 02/26/10 12:45 AM
Joined: Aug 2009
Posts: 132
L
lqz Offline OP
Forum Member
OP Offline
Forum Member
L
Joined: Aug 2009
Posts: 132
Thanks for your hint. In fact, some residues are not defined in the new topology file. So I copied some of the parameters from the old edition to the new edition. That caused some ambiguity, especially in the par file. I rearranged the order of the copied part and the default part in the par file. Now the warnings are gone. I only got following message when reading the rtf files:

CHARMM> OPEN READ UNIT 2 CARD NAME ./top_all22_prot-all.inp
VOPEN> Attempting to open::./top_all22_prot-all.inp::
OPNLGU> Unit 2 opened for READONLY access to /home/lxz79/charmm-test/./top_all22_prot-all.inp

CHARMM> READ RTF UNIT 2 CARD
MAINIO> Residue topology file being read from unit 2.
TITLE> *>>>>>>>>CHARMM22 ALL-HYDROGEN TOPOLOGY FILE FOR PROTEINS <<<<<<
TITLE> *>>>>> INCLUDES PHI, PSI CROSS TERM MAP (CMAP) CORRECTION <<<<<<<
TITLE> *>>>>>>>>>>>>>>>>>>>>>> DECEMBER, 2003 <<<<<<<<<<<<<<<<<<<<<<<<<<
TITLE> * ALL COMMENTS TO ADM JR. VIA THE CHARMM WEB SITE: www.CHARMM.ORG
TITLE> * PARAMETER SET DISCUSSION FORUM
TITLE> *

CHARMM> CLOSE UNIT 2
VCLOSE: Closing unit 2 with status "KEEP"

CHARMM>

CHARMM> !
CHARMM> !Read nucleo-acid parameters
CHARMM> !
CHARMM> OPEN READ UNIT 2 CARD NAME ./par_all22_prot-all.inp
VOPEN> Attempting to open::./par_all22_prot-all.inp::
OPNLGU> Unit 2 opened for READONLY access to /home/lxz79/charmm-test/./par_all22_prot-all.inp

CHARMM> READ PARAM UNIT 2 CARD

PARAMETER FILE BEING READ FROM UNIT 2
TITLE> *>>>>> CHARMM22 ALL-HYDROGEN PARAMETER FILE FOR PROTEINS <<<<<<<<
TITLE> *>>>>> INCLUDES PHI, PSI CROSS TERM MAP (CMAP) CORRECTION <<<<<<<
TITLE> *>>>>>>>>>>>>>>>>>>>>>> DECEMBER, 2003 <<<<<<<<<<<<<<<<<<<<<<<<<<
TITLE> * ALL COMMENTS TO ADM JR. VIA THE CHARMM WEB SITE: www.CHARMM.ORG
TITLE> * PARAMETER SET DISCUSSION FORUM
TITLE> *
PARMIO> NONBOND, HBOND lists and IMAGE atoms cleared.

The last line showed that NONBOND, HBOND lists and IMAGE atoms cleared. Is that related to the par file? thanks a lot.

Re: c36a4 topology file changed
lqz #23531 02/26/10 07:09 AM
Joined: Sep 2003
Posts: 4,797
Likes: 2
Forum Member
Online Content
Forum Member
Joined: Sep 2003
Posts: 4,797
Likes: 2
That all seems OK, but random manipulation of rtf/parameter files may still lead to unreliable results.


Lennart Nilsson
Karolinska Institutet
Stockholm, Sweden
Re: c36a4 topology file changed
lennart #23542 02/26/10 04:23 PM
Joined: Aug 2009
Posts: 132
L
lqz Offline OP
Forum Member
OP Offline
Forum Member
L
Joined: Aug 2009
Posts: 132
I only copied and pasted information that new topology files do not have but my proteins need. I do not know if there is other way to do it and make the results more trustable.

Re: c36a4 topology file changed
lqz #23544 02/26/10 04:31 PM
Joined: Sep 2003
Posts: 4,797
Likes: 2
Forum Member
Online Content
Forum Member
Joined: Sep 2003
Posts: 4,797
Likes: 2
Keep using the old files. Why do you need to use the new files?
Or find where the information is in the new set of files.
If the information is something that has been added locally it would have been a great help if you had told us about this in the first place.


Lennart Nilsson
Karolinska Institutet
Stockholm, Sweden
Re: c36a4 topology file changed
lennart #23550 02/26/10 06:46 PM
Joined: Dec 2005
Posts: 1,535
Forum Member
Offline
Forum Member
Joined: Dec 2005
Posts: 1,535
I think Lennart is on to something; this whole story just doesn't make sense. No changes have been made to the official protein topology file since the introduction of the CMAP term, which has been integrated in toppar/top_all22_prot.inp starting from C35. And those changes are additions, not deletions. And yes, I double-checked the c36a4 distribution. You must have been working with modified versions of those files that are internal to your lab (and hence outside of the scope of this forum).

Re: c36a4 topology file changed
Kenno #23554 02/26/10 08:13 PM
Joined: Aug 2009
Posts: 132
L
lqz Offline OP
Forum Member
OP Offline
Forum Member
L
Joined: Aug 2009
Posts: 132
You are right. I believe the extra residue definition part are internal to my lab only, for example, the gtp. Although I do not know where it originally came from. I just copied those parts to the new topology files since I could not find a better way to do.

Comparing the former and the new topology files, I could see that several parameters are updated. Since I am going to work on new projects and I believe the new charmm and topology files are better, I prefer to use new topology files.

Re: c36a4 topology file changed
lqz #23555 02/26/10 08:20 PM
Joined: Sep 2003
Posts: 4,797
Likes: 2
Forum Member
Online Content
Forum Member
Joined: Sep 2003
Posts: 4,797
Likes: 2
Unless your original files are very old I don't think there should be any significant parameter changes. How could they be only "slightly different from c36a4" if there are whole residues added locally?

Your habit of hiding (or providing misleading) information makes it very difficult to help you.


Lennart Nilsson
Karolinska Institutet
Stockholm, Sweden
Re: c36a4 topology file changed
lqz #23556 02/26/10 08:49 PM
Joined: Dec 2005
Posts: 1,535
Forum Member
Offline
Forum Member
Joined: Dec 2005
Posts: 1,535
Originally Posted By: lqz
Comparing the former and the new topology files, I could see that several parameters are updated. Since I am going to work on new projects and I believe the new charmm and topology files are better, I prefer to use new topology files.

When talking about the parameter file, you should say "parameter file", not "topology file".

Parameters were not updated with the addition of the CMAP term, only added. Either you're comparing against a really old (pre-2004) version of the parameter file, or people in your lab fiddled with the parameters. Either way, this is something you have to sort out with the people in your lab, not with the people on this forum.

Lastly, it cannot be guaranteed that the residues added by the people in your lab are compatible with the newer parameters.

Oh yeah, can a forum administrator move this thread to "parameter set discussion"? Thanks in advance! smile

Last edited by Kenno; 02/26/10 08:53 PM. Reason: Added two last sentences.
Page 1 of 2 1 2

Moderated by  alex, lennart, rmv 

Link Copied to Clipboard
Powered by UBB.threads™ PHP Forum Software 7.7.4
(Release build 20200307)
Responsive Width:

PHP: 5.6.33-0+deb8u1 Page Time: 0.023s Queries: 50 (0.012s) Memory: 1.0762 MB (Peak: 1.2752 MB) Data Comp: Off Server Time: 2020-10-31 18:32:45 UTC
Valid HTML 5 and Valid CSS