From: Ivan Gregoretti (ivangreg_at_gmail.com)
Date: Tue Feb 25 2014 - 07:35:49 CST

With the new version of psfgen from VMD 1.9.2 a35 I was able to
prepare the 3GRW structure for MD.

It run with NAMD 2.9 without problems.

Thank you all for helping. Closing the thread.

Ivan

Ivan Gregoretti, PhD
Bioinformatics

On Mon, Feb 24, 2014 at 8:43 PM, Tristan Croll <tristan.croll_at_qut.edu.au> wrote:
> Well, I'm glad to have learned something here.
>
> -----Original Message-----
> From: owner-vmd-l_at_ks.uiuc.edu [mailto:owner-vmd-l_at_ks.uiuc.edu] On Behalf Of Boris Steipe
> Sent: Tuesday, 25 February 2014 2:16 AM
> To: Ivan Gregoretti
> Cc: vmd-l list
> Subject: Re: vmd-l: psfgen and unknown residue types
>
> Sorry - I can't help you with these particulars. I was just responding to the notion that the 3GRW had misnumbered residues. They are not misnumbered. There is a perfectly good reason why some proteins - immunoglobulins especially - use non-sequential numbering schemes: this is to make residue numbers comparable within protein families that have indels in their sequence. Therefore such insert codes have been a part of the PDB format from the time that records were still written on punch-cards.
>
> If it is indeed the case that such insert codes break assumptions by autopsf, as Tristan said, then this is an autopsf bug.
>
> I hope this helps,
> Boris
>
>
>
>
> On 2014-02-24, at 10:51 AM, Ivan Gregoretti wrote:
>
>> Hi Boris,
>>
>> Would you mind showing us how you would process this structure in
>> preparation for NAMD?
>>
>> This is the content of the scripts I wrote and executed:
>>
>> Script 1
>> mol new 3GRW_p.pdb
>> set A [atomselect 0 "chain A"]
>> $A writepdb 3GRW_p_A.pdb
>> exit
>>
>> Script 2
>> mol new 3GRW_p.pdb
>> set L [atomselect 0 "chain L"]
>> $L writepdb 3GRW_p_L.pdb
>> exit
>>
>> Script 3
>> mol new 3GRW_p.pdb
>> set H [atomselect 0 "chain H"]
>> $H writepdb 3GRW_p_H.pdb
>> exit
>>
>> Script 4
>> package require psfgen
>> topology top_all27_prot_lipid.inp
>> pdbalias residue HIS HSE
>> pdbalias atom ILE CD1 CD
>> segment A {pdb 3GRW_p_A.pdb}
>> coordpdb 3GRW_p_A.pdb A
>> guesscoord
>> writepdb 3GRW_p_A_hy.pdb
>> writepsf 3GRW_p_A_hy.psf
>> exit
>>
>> Script 5
>> package require psfgen
>> topology top_all27_prot_lipid.inp
>> pdbalias residue HIS HSE
>> pdbalias atom ILE CD1 CD
>> segment L {pdb 3GRW_p_L.pdb}
>> coordpdb 3GRW_p_L.pdb L
>> guesscoord
>> writepdb 3GRW_p_L_hy.pdb
>> writepsf 3GRW_p_L_hy.psf
>> exit
>>
>> Script 6
>> package require psfgen
>> topology top_all27_prot_lipid.inp
>> pdbalias residue HIS HSE
>> pdbalias atom ILE CD1 CD
>> segment H {pdb 3GRW_p_H.pdb}
>> coordpdb 3GRW_p_H.pdb H
>> guesscoord
>> writepdb 3GRW_p_H_hy.pdb
>> writepsf 3GRW_p_H_hy.psf
>> exit
>>
>> Script 7 (putting the three peptides together again)
>> package require psfgen
>> readpsf 3GRW_p_A_hy.psf
>> coordpdb 3GRW_p_A_hy.pdb
>> readpsf 3GRW_p_L_hy.psf
>> coordpdb 3GRW_p_L_hy.pdb
>> readpsf 3GRW_p_H_hy.psf
>> coordpdb 3GRW_p_H_hy.pdb
>> writepsf 3GRW_p_X_hy_comb.psf
>> writepdb 3GRW_p_X_hy_comb.pdb
>> exit
>>
>> Do you reckon that my script 6 at least need a modification?
>> By the way, I am using VMD 1.9.1.
>>
>> Thank you,
>>
>> Ivan
>>
>>
>>
>>
>> Ivan Gregoretti, PhD
>> Bioinformatics
>>
>>
>>
>> On Sun, Feb 23, 2014 at 10:19 PM, Boris Steipe <boris.steipe_at_utoronto.ca> wrote:
>>> There is nothing at all wrong with 3GRW, the problem is a bug in whatever processing is being done here that ignore the insert code which is part of a unique residue "number" in PDB files. The residues are "100 ", "100A", "100B" .. etc. One can't simply treat PDB residue numbers as unique, consecutive integers.
>>>
>>> Cheers,
>>> Boris
>>>
>>>
>>>
>>> On 2014-02-23, at 7:56 PM, Ivan Gregoretti wrote:
>>>
>>>> Well, I picked 3GRW from RCSB somehow randomly. I wanted something
>>>> more complex than the classic tutorial on ubiquitin molecular
>>>> dynamics.
>>>>
>>>> I see now that I bit something bigger than I could chew. A
>>>> misnumbering of residues in the public 3GRW.pdb file is not something
>>>> that crossed my mind. I'll start there, numbering properly.
>>>>
>>>> By the way, I am not using AutoPSF but tcl scripts that I write myself
>>>> looking at documentation available on web.
>>>>
>>>> The generosity with your time is highly appreciated. You probably
>>>> saved me a couple days of frustration. Analysis of Next Generation
>>>> Sequencing datasets is my forte. I hope I can reciprocate one day.
>>>>
>>>> Ivan
>>>>
>>>>
>>>>
>>>>
>>>> Ivan Gregoretti, PhD
>>>> Bioinformatics
>>>>
>>>>
>>>>
>>>> On Sun, Feb 23, 2014 at 7:13 PM, Tristan Croll <tristan.croll_at_qut.edu.au> wrote:
>>>>> Hi Ivan,
>>>>>
>>>>> I'm feeling a little generous today, so I looked more closely at this. Your problem has nothing to do with the NAG residue - I'm quite sure you can safely delete this. Your problem is that for some reason 3grw has 11 residues in chain H all labelled as residue 100. AutoPSF doesn't know what to do with these, so deletes all but one - which of course will lead to all sorts of strife. Renumbering the residues should fix the problem.
>>>>>
>>>>> Cheers,
>>>>>
>>>>> Tristan
>>>>>
>>>>> -----Original Message-----
>>>>> From: Ivan Gregoretti [mailto:ivangreg_at_gmail.com]
>>>>> Sent: Monday, 24 February 2014 9:39 AM
>>>>> To: Tristan Croll; vmd-l list
>>>>> Subject: Re: vmd-l: psfgen and unknown residue types
>>>>>
>>>>> Hi Tristan and everybody,
>>>>>
>>>>> NAG (N-acetyl glucosamine) seems to be tricky case.
>>>>>
>>>>> I went to http://mackerell.umaryland.edu and downloaded
>>>>> http://mackerell.umaryland.edu/download.php?filename=CHARMM_ff_params_files/toppar_c36_dec13.tgz
>>>>>
>>>>> Then I decompressed the tar ball and did
>>>>>
>>>>> find ./toppar -name '*' | sort | tr \\n \\0 | xargs -0 grep --colour 'NAG'
>>>>> ./toppar/non_charmm/par_bms_dec03.inp:CK NAG 340.0 1.410
>>>>> !G, par_a4 JWK 9/30/91
>>>>> ./toppar/non_charmm/par_bms_dec03.inp:CA NAG 400.0 1.400
>>>>> !G, JWK, adm jr., 7/24/91
>>>>> ./toppar/non_charmm/par_bms_dec03.inp:CE NAG 350.0 1.365
>>>>> !Inosine, adm jr., 2/94
>>>>> ./toppar/non_charmm/par_bms_dec03.inp:HB NAG 471.0 1.01
>>>>> !G, JWK, par_a12 9/30/91
>>>>> ./toppar/non_charmm/par_bms_dec03.inp:CTT NAG 400.0 1.46
>>>>> !DRL for 1methylguanine
>>>>> ./toppar/non_charmm/par_bms_dec03.inp:CB CK NAG 67.0 109.4
>>>>> !G, JWK, adm jr., 7/24/91
>>>>> ./toppar/non_charmm/par_bms_dec03.inp:!!!NAG CK OK 100.0
>>>>> 125.0 !G, par_a4 adm jr., 10/2/91
>>>>> ./toppar/non_charmm/par_bms_dec03.inp:NAG CK OKG 100.0 125.0 !G, DRL
>>>>> ./toppar/non_charmm/par_bms_dec03.inp:NAG CK S 100.0
>>>>> 121.45 ! 80.0 120.6 !TG, DRL, B3LYP/6-31+G** (119.45)
>>>>> ./toppar/non_charmm/par_bms_dec03.inp:NAN CA NAG 70.0 114.8
>>>>> 20. 2.30 !G, JWK, adm jr., 7/24/91
>>>>> ./toppar/non_charmm/par_bms_dec03.inp:NAG CA NCA 70.0 122.9
>>>>> !G, JWK par_49, par_a11 9/30/91
>>>>> ./toppar/non_charmm/par_bms_dec03.inp:HAR CE NAG 35.0
>>>>> 114.75 !Inosine, adm jr., 2/94
>>>>> ./toppar/non_charmm/par_bms_dec03.inp:CK NAG CA 125.0 128.7
>>>>> !G, JWK, adm jr., 7/24/91
>>>>> ./toppar/non_charmm/par_bms_dec03.inp:CK NAG CE 125.0 129.7
>>>>> !Inosine, adm jr., 2/94
>>>>> ./toppar/non_charmm/par_bms_dec03.inp:CK NAG HB 40.5 114.3
>>>>> !G, adm jr., 7/24/91
>>>>> ./toppar/non_charmm/par_bms_dec03.inp:CA NAG HB 40.5 117.0
>>>>> !G par_49, adm jr., 7/24/91
>>>>> ./toppar/non_charmm/par_bms_dec03.inp:CE NAG HB 40.5 116.0
>>>>> !Inosine, adm jr., 2/94
>>>>> ./toppar/non_charmm/par_bms_dec03.inp:CK NAG CTT 40.5 114.3
>>>>> !G, DRL for 1-methylguanine
>>>>> ./toppar/non_charmm/par_bms_dec03.inp:CA NAG CTT 40.5 117.0
>>>>> !G, DRL for 1-methylguanine
>>>>> ./toppar/non_charmm/par_bms_dec03.inp:HAT CTT NAG 43.0 111.0
>>>>> !G, DRL for 1-methylguanine
>>>>> ./toppar/non_charmm/par_bms_dec03.inp:!!!OK CK NAG HB 3.0
>>>>> 2 180.0 ! PAR_36, adm jr., 7/24/91
>>>>> ./toppar/non_charmm/par_bms_dec03.inp:OKG CK NAG HB 3.0
>>>>> 2 180.0 !G, DRL
>>>>> ./toppar/non_charmm/par_bms_dec03.inp:NAG CK CB CB 6.0
>>>>> 2 180.0 ! par_49
>>>>> ./toppar/non_charmm/par_bms_dec03.inp:CB CK NAG CA 6.0
>>>>> 2 180.0 ! par_49
>>>>> ./toppar/non_charmm/par_bms_dec03.inp:CK NAG CA NCA 6.0
>>>>> 2 180.0 ! par_49
>>>>> ./toppar/non_charmm/par_bms_dec03.inp:NAG CA NCA CB 6.0
>>>>> 2 180.0 ! par_49
>>>>> ./toppar/non_charmm/par_bms_dec03.inp:NAG CA NAN HNA 2.0
>>>>> 2 180.0 ! par_49
>>>>> ./toppar/non_charmm/par_bms_dec03.inp:HAR CE NAG HB 1.5 2 180.0
>>>>> ./toppar/non_charmm/par_bms_dec03.inp:HAR CE NAG CK 1.5 2 180.0
>>>>> ./toppar/non_charmm/par_bms_dec03.inp:X CK NAG X 0.9
>>>>> 2 180.0 ! par_36
>>>>> ./toppar/non_charmm/par_bms_dec03.inp:X CA NAG X 1.0
>>>>> 2 180.0 ! par_49
>>>>> ./toppar/non_charmm/par_bms_dec03.inp:CA NAG CTT HAT 0.19 3 0.0
>>>>> !DRL for 1-methylguanine
>>>>> ./toppar/non_charmm/par_bms_dec03.inp:CK NAG CTT HAT 0.19 3 0.0
>>>>> !DRL for 1-methylguanine
>>>>> ./toppar/non_charmm/par_bms_dec03.inp:HB X X NAG 0.3
>>>>> 0 0.0 !G, JWK par_15
>>>>> ./toppar/non_charmm/par_bms_dec03.inp:HB CE CK NAG 0.8
>>>>> 0 0.0 !Inosine, adm jr., 2/94
>>>>> ./toppar/non_charmm/par_bms_dec03.inp:CTT X X NAG 20.0
>>>>> 0 0.0 !G, DRL for 1-methylguanine
>>>>> ./toppar/non_charmm/par_bms_dec03.inp:NAG 0.0 -0.20 1.85
>>>>> ./toppar/non_charmm/top_bms_dec03.inp:MASS 20 NAG 14.007000 !
>>>>> Nucleic acid protonated ring nitrogen, gua N1
>>>>> ./toppar/non_charmm/top_bms_dec03.inp:ATOM N1 NAG -0.5053 !-0.59
>>>>> ./toppar/non_charmm/top_bms_dec03.inp:ATOM N1 NAG -0.4787
>>>>> ./toppar/non_charmm/top_bms_dec03.inp:ATOM N1 NAG -0.3635
>>>>> ./toppar/non_charmm/top_bms_dec03.inp:ATOM N1 NAG -0.1325 !-0.5053
>>>>> ./toppar/param19.inp:FE NR 65.0 1.98! FROM NAGAI ET AL(1980)
>>>>>
>>>>> or, simply
>>>>>
>>>>> find ./toppar -name '*' | sort | tr \\n \\0 | xargs -0 grep -l --colour 'NAG'
>>>>> ./toppar/non_charmm/par_bms_dec03.inp
>>>>> ./toppar/non_charmm/top_bms_dec03.inp
>>>>> ./toppar/param19.inp
>>>>>
>>>>> The only matches to NAG are in the non-CHARMM topologies and
>>>>> parameters. The accompanying ./toppar/00toppar_file_format.txt says
>>>>>
>>>>> "subdirectory non_charmm: Contains toppar files for AMBER,
>>>>> Bristol-Myers Squibb (BMS) and OPLS force fields along with stream
>>>>> files for a variety of water models. These files have been tested
>>>>> to the extent that they may be considered reliable representations of
>>>>> the original force fields, though potentially not exact
>>>>> representations. These files are NOT maintained and, thus, use at
>>>>> your own risk. See the 00readme files and note that AMBER requires a
>>>>> special version of CHARMM as described in the 00readme file."
>>>>>
>>>>> So, it seems that there is no CHARMM for NAG, not even in CHARMM 36.
>>>>>
>>>>> I tried running namd excluding the NAG. It did run but after a few
>>>>> steps namd's log reports:
>>>>>
>>>>> "...ERROR: Constraint failure in RATTLE algorithm for atom 86!
>>>>> ERROR: Constraint failure; simulation has become unstable."
>>>>>
>>>>> I reckon that NAG is needed to keep the three peptide chains together.
>>>>>
>>>>> More ideas?
>>>>>
>>>>> Thank you,
>>>>>
>>>>> Ivan
>>>>>
>>>>>
>>>>>
>>>>>
>>>>> Ivan Gregoretti, PhD
>>>>> Bioinformatics
>>>>>
>>>>>
>>>>>
>>>>> On Sun, Feb 23, 2014 at 3:41 AM, Tristan Croll <tristan.croll_at_qut.edu.au> wrote:
>>>>>> "NAG" is N-acetyl glucosamine - probably a glycan linked to an asparagine residue. If you want to keep this in your models, you'll have to use a forcefield that supports glycans, such as CHARMM-36. "HOH" is, of course, water. If all you want is a standard protein simulation, just delete all segnames not starting with P during AutoPSF.
>>>>>>
>>>>>>
>>>>>>
>>>>>> Tristan Croll
>>>>>> Lecturer
>>>>>> Faculty of Science and Technology
>>>>>> Institute of Health and Biomedical Engineering
>>>>>> Queensland University of Technology
>>>>>> 60 Musk Ave
>>>>>> Kelvin Grove QLD 4059 Australia
>>>>>> +61 7 3138 6443
>>>>>>
>>>>>> This email and its attachments (if any) contain confidential information intended for use by the addressee and may be privileged. We do not waive any confidentiality, privilege or copyright associated with the email or the attachments. If you are not the intended addressee, you must not use, transmit, disclose or copy the email or any attachments. If you receive this email by mistake, please notify the sender immediately and delete the original email.
>>>>>>
>>>>>>
>>>>>>
>>>>>>> On 23 Feb 2014, at 10:26 am, "Ivan Gregoretti" <ivangreg_at_gmail.com> wrote:
>>>>>>>
>>>>>>> When running psfgen I get two complains for the "A" chain of 3GRW:
>>>>>>>
>>>>>>>
>>>>>>> ...
>>>>>>> unknown residue type NAG
>>>>>>> unknown residue type HOH
>>>>>>> ...
>>>>>>> Info: generating structure...psfgen) unknown residue type NAG
>>>>>>> failed!
>>>>>>> ERROR: failed on end of segment
>>>>>>> MOLECULE DESTROYED BY FATAL ERROR! Use resetpsf to start over.
>>>>>>> ...
>>>>>>>
>>>>>>>
>>>>>>> Could anybody point me to a document to read and figure out a solution?
>>>>>>>
>>>>>>> Thank you,
>>>>>>>
>>>>>>> Ivan
>>>>>>>
>>>>>>>
>>>>>>>
>>>>>>>
>>>>>>> Ivan Gregoretti, PhD
>>>>>>> Bioinformatics
>>>>
>>>
>
>