I build a system which contains RNA, protein and a membrane with Cholesterol so i need to read in the parameter files for all of the four components (NA, PROT, LIPID, CHOL).
I checked each structure by its own and made sure each works. But I failed at the point where I want to put the system together. I couldn't manage to read all the required rtf/prm files.
I tried different combinations and orders but no success. See an example below:
Code:
! Read topology and parameter files
open read card unit 10 name para/top_all27_prot_na.rtf
read rtf card unit 10
open read card unit 20 name para/par_all27_prot_na.prm
read para card unit 20
open read card unit 10 name para/top_all27_lipid.rtf
read rtf card append unit 10
open read card unit 20 name para/par_all27_lipid.prm
read para card append unit 20
stream para/toppar_all27_lipid_cholesterol.str
Before I start to post all possible errors I collected I would like to ask if this is the proper way of reading multiple rft/prm files. If not, please let me know how to read all the needed parameters (NA, PROT, LIPID, CHOLESTEROL)
A good question for the Parameter forum, and a topic which may have been discussed there before. There's also a post there about the new "all36_lipid" files, which I recommend over either "all27_lipid" (worst choice) or "all32_lipid" (alkanes better).
In brief, the above won't work with the current structure of the RTF and PARAM files (a revision is planned). You must manually combine the RTF files, taking care to avoid duplicate atom numbers and names in the MASS statements, and duplicate residues. I suggest adding the "all36_lipid" files to the "all27_prot_na" files as the best and easiest pathway. For the PARAM files, you should combine each section, i.e. add lipid BOND terms to existing BOND section, etc.
You're in the wrong place. This is the CHARMM forum!
To solve your problem, you have to renumber the MASS statements in the topology file, copy-paste them into a new "ATOM" section in the parameter file, and read the resulting parameter files with the flex keyword. Please read this discussion thread carefully (including the second page) for a discussion on the subject.
We're currently in the process of reorganizing the CHARMM force field distribution. All the force fields will have non-overlapping MASS numbers and will be compatible with the flexible parameter reader. That should make these kind of annoyances a thing of the past. We're planning to release the new CHARMM force field distribution next week.
Thanks for your replies, I appreciate it very much.
I had the apprehension that it would be the way you wrote.
Since my parent simulation was run with all27_lipid I rather stick with this, shouldn't I? If not, I'll see if we can purchase the latest CHARMM ;-) version when it got release to get my hands on the new forcefield files and avoid messing around with these.
Meanwhile I'll try to combine all27_prot_na and all27_lipid.
The force field files can be downloaded by anyone from the MacKerell lab site.
Ah, good to know, thanks for the info.
I realized that I used top_all27_prot_lipid to setup the membrane and since O3 referes to OHL in toppar_all27_lipid_cholesterol and OHL is type 165 in top_all27_prot_lipid I think I have found the mistake.
I'll try to append the all27_na to all27_prot_lipid and regenerate the RNA psf from the PDB (or change types... python will come i handy here)
So please ignore the stuff below - for the moment -
Why don't you just edit your post and cut-and-paste the stuff "to be ignored" to a text file on your desktop? That would prevent people from stumbling across it while using the search function and getting confused. It's a question of keeping the noise down. But do it fast, because the "edit" function gets automatically disabled after a number of hours.
But do it fast, because the "edit" function gets automatically disabled after a number of hours.
Done
Sadly the above didn't work. I merged the all27_na with all27_prot_lipid, removed all duplicates but when re-creating my PSF from a PDB I get some errors which I can't identify but which for sure result from an error in my prm/rtf files.
I pasted the prm/rtf files here: http://paste.pocoo.org/show/231347/ (scroll down to line 5371 for top_all27_prot_lipid_na) or find them attached to this post.
The part where I added the NA stuff is marked with !!!!NA_START!!!! and !!!!NA_STOP!!!!
[...]
PARAMETER FILE BEING READ FROM UNIT 20
TITLE> *>CHARMM22 ALL-HYDROGEN PARAMETER FILE FOR PROTEINS AND LIPIDS <<
TITLE> *>>>>> INCLUDES PHI, PSI CROSS TERM MAP (CMAP) CORRECTION <<<<<<<
TITLE> *>>>>>>>>>>>>>>>>>>>>>> JULY, 2003 <<<<<<<<<<<<<<<<<<<<<<<<<<
TITLE> * ALL COMMENTS TO ADM JR. VIA THE CHARMM WEB SITE: WWW.CHARMM.ORG
TITLE> * PARAMETER SET DISCUSSION FORUM
TITLE> *
WARNING from DECODI -- Zero length string being converted to 0
PARRDR> WARNING: ATOMS IN PHI1 (CMAP) HN2 X X NN2 DONT EXIST
[...]
PARRDR> NOTE: atom type "HT " is removed from previous group
PARRDR> NOTE: atom type "OT " is removed from previous group
***** WARNING ***** PARRDR no nonbond parameters for atom type: 250 POLO
NO nonbond interactions will be computed for this atom type.
PARRDR: IATVDW 38
PARRDR: NATVDW 192
PARRDR: NBC 20 21 32 33 39 40 37 29 30 31
34 35 25 26 36 27 38 41 22 23
[...]
PARRDR: NATC 334
PARRDR: ITC 27 32 28 122 33 31 34 35 36 37
29 30 39 40 41 85 86 0 0 1
[...]
***** LEVEL -3 WARNING FROM <PARRDR> *****
***** Null nonbond group found. Redo.
******************************************
BOMLEV ( 0) IS REACHED - TERMINATING. WRNLEV IS 5
this is my input file (just in case):
Code:
*Regenerate RNA because it was build with the standard parameter
*files and thus the MASS (type) of atoms do not match
*
open read card unit 10 name ../para/top_all27_prot_lipid_na.rtf
read rtf card unit 10
open read card unit 20 name ../para/par_all27_prot_lipid_na.prm
read para card unit 20
open read card unit 10 name ../input/rna.pdb
read sequence pdb unit 10
generate RNAA setup warn first 5TER last 3TER
open read card unit 10 name ../input/rna.pdb
read coor pdb unit 10 resid
!Print heavy atoms with unknown coordinates
coor print sele ( .not. INIT ) .and. ( .not. hydrogen ) end
ic param
ic build
prnlev 0
hbuild
prnlev 5
open write card unit 10 name rna.pdb
write coor pdb unit 10
open write card unit 10 name rna.crd
write coor unit 10 card
open write unit 10 card name rna.psf
write psf unit 10 card
open write unit 10 card name rna.xplor.psf
write psf xplo unit 10 card
stop
It's just a copy-paste error. In your parameter file, you have an "IMPROPER" section, then a "CMAP" section, and then a bunch of loose improper definitions that CHARMM cannot process because they're in the CMAP section...
It probably would have taken you less time to analyze the error message and find the problem by yourself than to dump all this stuff on the CHARMM forums. Not to mention the time we would have saved, and the missed opportunity to practice your problem-solving skills. More on that topic in my problem solving 101 rant.
