November 2001 - modeller_usage

Error using MALIGN3D
by Michael Buck 24 Jan '02

24 Jan '02

MALIGN3D Top files which worked well using Modeler 4 are giving me this error : fndatmi_284E> Beginning atom index for a residue out of bounds: 0 0 recover____E> ERROR_STATUS >= STOP_ON_ERROR: 1 1 I am using the following script # MALIGN3D READ_ALIGNMENT FILE = 'pair.aa' SET ATOM_FILES = '1.pdb' '2.pdb' SET FIT_ATOMS = 'CA' MALIGN3D GAP_PENALTIES_3D= 0 2.0, WRITE_FIT = on, WRITE_WHOLE_PDB = off If anybody has seen this problem, please help. Thank you Michael Buck ************************************************ Michael Buck NCSU Genetics mjbuck(a)unity.ncsu.edu Phone (919)515-5759 Fax (919)515-3355 http://www4.ncsu.edu/~mjbuck *************************************************

2 1

Regarding obtaining dimer form
by R Senthil Kumar 29 Nov '01

29 Nov '01

Dear modeller users, I am modelling a protein which is of homodimer in nature and this protein matched with the same family protein which is also of homodimer. But matching with the only one chain(of two chains).So , i got a model in a monomer form.How i have to use the modeller to get the replicate of other monomer so my model will be of homodimer in nature.Please suggest me ideas to overcome this problem.Thanks in advance.Expecting reply from your people. ******************************** R.Senthil kumar Junior Research Fellow(JRF), C/O Dr.Akash Ranjan, Computational & Functional Genomics(CFG), Centre for DNA Fingerprinting & Diagnostics(CDFD), ECIL road, Nacharam, Hyderabad - 500 076. India. E-mail:skumar@www.cdfd.org.in senkumar7(a)yahoo.com Ph.no:040-7151344-EXTN(1304)--->Lab (2300)--->Hostel ********************************

2 2

Re: Regarding obtaining dimer form
by Eric Feyfant 29 Nov '01

29 Nov '01

You got only one match in your sequence search because you ran a search for only one chain. But because your template and target are HOMOdimer use as alignment file in your .top script this kind of alignment: >P1;Template StructureX::::::: AAAAAAAAAAAAAAA/AAAAAAAAAAAAAAA >P1;Target Sequence:::::: BBBBBBBBBBBBBB/BBBBBBBBBBBBBBBB good luck, Eric -------------------------------------------------------- Eric Feyfant, Ph.D. Department of Biological Chemistry WYETH Research 85 bolton St. Cambridge ,MA 02140 tel: (617) 665 8987 fax:(617) 665 8993 email: efeyfant(a)genetics.com >>> R Senthil Kumar <skumar(a)www.cdfd.org.in> 11/29 7:11 PM >>> Dear Eric, Thank you very much for your reply.Eric i will tell my problem clearly. I am having a protein of 274 amino acids.When i searched for the template for this protein it picked up the template which is of homodimer in nature. i.e, it is having aaaaaaaaaaaaaa/aaaaaaaaaaaaaaaaa(two chains).My protein is also belong to the same family of template.In this family homodimer is of active form. My protein matching with only to the first chain. If i go ahead with this alignment i will get my peotein in monomeric form. My question is 1.Whether i can go ahead and get my protein in monomeric form and after that i can do any trick to convert it into homodimer form-->how to do this. 2.else in the alignment itself can i do any trick so that i can get homodimer model of my protein--->how to do this. If you tell me it will be of very helpful to me.Expecting reply from you. from, Senthil kumar.R R.Senthil Kumar, Junior Research Fellow(JRF), c/o Dr.Akash Ranjan, Computational & Functional Genomics Group, Centre For DNA Fingerprinting(CDFD), Hyderabad-500 076 Ph:040-7151344-Extn(Lab:1304) (Hostel:2300) E-mail:skumar@www.cdfd.org.in On Wed, 28 Nov 2001, Eric Feyfant wrote: > Dear Senthil; > > I don't understand something you said: > > you have an homodimer to model: seq: AAAAAAAAAAAAAAA/AAAAAAAAAAAA > You got a template that is an other homodimer so with sequence like: BBBBBBBBBBBB/BBBBBBBBBBBBB (/ separate the 2 chains). > Now you say that you got a match for only one chain???? > > if you can align 1 chain you should be able to align the 2nd chain since in the homodimer the 2 chains are identical? > So the alignment should look like : > > AAAAAAAAAAAAAA/AAAAAAAAAAAAAA > BBBBBBBBBBBBBB/BBBBBBBBBBBBBB > > Now if you know the binding site between the 2 chains are different (not only in sequence but also the binding site location) the situation is more complicate and I will suggest you to use biological data and do the docking manually or using docking software like ftdock where you can specify known interaction. > > For any further question, I will be happy to help you. > > Eric > > -------------------------------------------------------- > Eric Feyfant, Ph.D. > Department of Biological Chemistry > WYETH Research > 85 bolton St. > Cambridge ,MA 02140 > > tel: (617) 665 8987 > fax:(617) 665 8993 > email: efeyfant(a)genetics.com > > >>> R Senthil Kumar <skumar(a)www.cdfd.org.in> 11/28 9:06 PM >>> > > Dear modeller users, > > I am modelling a protein which is of homodimer in nature and this protein > matched with the same family protein which is also of homodimer. > > But matching with the only one chain(of two chains).So , i got a model in > a monomer form.How i have to use the modeller to get the replicate of > other monomer so my model will be of homodimer in nature.Please suggest me > ideas to overcome this problem.Thanks in advance.Expecting reply from your > people. > > ******************************** > R.Senthil kumar > Junior Research Fellow(JRF), > C/O Dr.Akash Ranjan, > Computational & Functional Genomics(CFG), > Centre for DNA Fingerprinting & Diagnostics(CDFD), > ECIL road, Nacharam, > Hyderabad - 500 076. > India. > > E-mail:skumar@www.cdfd.org.in > senkumar7(a)yahoo.com > > Ph.no:040-7151344-EXTN(1304)--->Lab > (2300)--->Hostel > > ******************************** > >

1 0

Re: Regarding obtaining dimer form
by Hafsa Ahsan 29 Nov '01

29 Nov '01

If the template is also a dimer and you say it is a homodimer then it means that the sequence should also be the same. Your target is also a homodimer which means here too the chains have similar amino acid sequence. If one chain of the target is homologous in sequence with the respective chain of the template then the other chain of the target should also be similar to the respective chain of the template. Here you do not have to do any modification in the script file or elsewhere, just align the two chains of the target with the two chains of the template and the modeller will automatically produce the dimer model, as Dr Sanchez has pointed out in his mail. Make sure that the alignment file is of correct format and the chains are separated by / sign. regards asmat ---------- > From: R Senthil Kumar <skumar(a)www.cdfd.org.in> > To: Roberto Sanchez <Roberto.Sanchez(a)physbio.mssm.edu> > Cc: modeller_usage(a)salilab.org > Subject: Re: Regarding obtaining dimer form > Date: Friday, November 30, 2001 5:29 AM > > > > Dear friends, > > > Thank you very much for your replies.First i will tell my problem > clearly. > > > I am having a protein of 274 amino acids.When i searched for the template > for this protein it picked up the template which is of homodimer in > nature. i.e, it is having > > aaaaaaaaaaaaaa/aaaaaaaaaaaaaaaaa(two chains).My protein is also belong to > the same family of template.In this family homodimer is of active form. > > My protein matching with only to the first chain. > > If i go ahead with this alignment i will get my peotein in monomeric form. > > My question is > > > 1.Whether i can go ahead and get my protein in monomeric form and after > that i can do any trick to convert it into homodimer form-->how to do > this. > > > 2.else in the alignment itself can i do any trick so that i can get > homodimer model of my protein--->how to do this. or in the script file. > > If you tell me it will be of very helpful to me.Expecting reply from you. > > from, > Senthil kumar.R >

1 0

Regarding modeller usage
by R Senthil Kumar 27 Nov '01

27 Nov '01

Dear modeller users, After i am getting the initial model using modeller if any bad regions are there(loop regions) i am using the following module for refinement. INCLUDE SET SEQUENCE = 'topmodel' SET MODEL = 'topmodel.pdb' SET STARTING_MODEL = 1 SET ENDING_MODEL = 2 CALL ROUTINE = 'loop' SUBROUTINE ROUTINE = 'select_atoms' PICK_ATOMS SELECTION_SEGMENT = '0:' '3:', SELECTION_STATUS ='initialize' RETURN END_SUBROUTINE Actually i want to know whether the following module does the refinement of the loop or the loop building with some other loop.Please tell what is the exact use of this module.Expecting for your reply. Yours friendly, Senthil kumar.R

2 1

Re: Crash in running the SEQUENCE_SRCH
by Tsu-Sheng Wang 22 Nov '01

22 Nov '01

Dears: I already tried the modeller6a in the RedHat 6.2. However, it still stoped in some step. I mailed the sequence and the log files to you. friendly, Jack ----- Original Message ----- From: "Andrej Sali" <sali(a)mail.rockefeller.edu> To: "Azat Badretdinov" <azat(a)accelrys.com>; "Tsu-Sheng Wang" <g894284(a)life.nthu.edu.tw> Cc: <modeller_usage(a)salilab.org> Sent: Thursday, November 15, 2001 12:33 AM Subject: RE: Crash in running the SEQUENCE_SRCH if still a problem, would you please try it with the pre-release of modeller 6 (ask bozidar(a)rockefeller.edu) if still a problem, would you please email us the input files. thanks, andrej -----Original Message----- From: owner-modeller_usage(a)salilab.org [mailto:owner-modeller_usage@salilab.org]On Behalf Of Azat Badretdinov Sent: Wednesday, November 14, 2001 9:56 AM To: Tsu-Sheng Wang Cc: modeller_usage(a)salilab.org Subject: Re: Crash in running the SEQUENCE_SRCH At 03:38 PM 11/14/01 +0800, Tsu-Sheng Wang wrote: >Dears: > > When I run below program, it will be crash. > >INCLUDE >SET OUTPUT_CONTROL = 1 1 1 1 1 >SET ATOM_FILES_DIRECTORY = '/usr/local/lib/pdb' >SET OUTPUT = 'SHORT' # 'LONG' >SET SIGNIF_CUTOFF = 8 3 >SET SEARCH_RANDOMIZATIONS = 10 >SET STOP_ON_ERROR = 2 >CALL ROUTINE = 'sequence_srch', SEGFILE = 'ny10.chn', SEQUENCE = 'ny10' > > I want to use this program to get the alignment and using the alignment >to get the template sequence and get the model by homology modelling. > The ny10.chn is one chain protein with 68 a.a. , and I set the >ATOM_FILES_DIRECTORY is the all PDB files in my server. This program will >be crash after abour the 45 min. I don't know what's wrong. My server is >RedHat Linux 7.1 with PIII 1G cpu and 512Mb ram. 7.1 is notorious for its bugs. It crashes randomly blastpgp jobs too at certain conditions. I would advise you to reproduce it on 6.2. >Your friendly, >Jack Wang > > -- Azat Badretdinov, Accelrys

1 0

Regarding modelling
by R Senthil Kumar 16 Nov '01

16 Nov '01

Dear modellers, I have started my Ph.D in the field of Molecular modelling and drug design.So,i need some comments about my modelling procedure.Please verify the modelling protocol that i am following to build the model and post-modelling procedures and give some good comments. 1. Obtaining the models with the help of modeller 2.Choose the model that is having lowest objective function. 3.Verify the model using Verify3D- program. 4.The regions that shows below the threshold value.Do the loop modelling and try to improve that region. 5.After improving the regions check it with also PORCHECK results for stereochemical evaluation. 6.Do the energy minimization with 50 cycles with fixing backbone and without charges.Then release the backbone and without charges for 50 cycles.Then with charges for 50 cycles. 7. Then verify it with PROSAII energy verification program. Expecting for your reply.Thanks in advance. Yours friendly, Senthil kumar.R

1 0

Crash in running the SEQUENCE_SRCH
by Tsu-Sheng Wang 14 Nov '01

14 Nov '01

Dears: When I run below program, it will be crash. INCLUDE SET OUTPUT_CONTROL = 1 1 1 1 1 SET ATOM_FILES_DIRECTORY = '/usr/local/lib/pdb' SET OUTPUT = 'SHORT' # 'LONG' SET SIGNIF_CUTOFF = 8 3 SET SEARCH_RANDOMIZATIONS = 10 SET STOP_ON_ERROR = 2 CALL ROUTINE = 'sequence_srch', SEGFILE = 'ny10.chn', SEQUENCE = 'ny10' I want to use this program to get the alignment and using the alignment to get the template sequence and get the model by homology modelling. The ny10.chn is one chain protein with 68 a.a. , and I set the ATOM_FILES_DIRECTORY is the all PDB files in my server. This program will be crash after abour the 45 min. I don't know what's wrong. My server is RedHat Linux 7.1 with PIII 1G cpu and 512Mb ram. Your friendly, Jack Wang

3 2

Regarding energy minimization using modeller
by R Senthil Kumar 14 Nov '01

14 Nov '01

Dear friends, I got a initial model for my protein using modeller.Then i went for loop modelling for some of the regions that showed below the threshold value in verfiy3D program. After the loop modelling i want to do energy minimization using optimize.top file(is this top file can be used for energy minimization). The following is the top file i used for energy minimization. Example for: OPTIMIZE, SWITCH_TRACE # This will optimize stereochemistry of a given model, including # non-bonded contacts. READ_TOPOLOGY FILE = '$(LIB)/top_heav.lib' READ_PARAMETERS FILE ='$(LIB)/par.lib' READ_MODEL FILE = 'topmodel6' SEQUENCE_TO_ALI ATOM_FILES= 'topmodel6', ALIGN_CODES = 'topmodel6' SEQUENCE_TO_ALI ADD_SEQUENCE =on, ATOM_FILES = ATOM_FILES 'topmodel6.ini',; ALIGN_CODES = ALIGN_CODES 'topmodel6-ini' GENERATE_TOPOLOGY SEQUENCE = 'topmodel6-ini' TRANSFER_XYZ BUILD_MODEL INITIALIZE_XYZ = off WRITE_MODEL FILE = 'topmodel6.ini' # Generate the restraints: MAKE_RESTRAINTS RESTRAINT_TYPE = 'stereo' WRITE_RESTRAINTS FILE = 'topmodel6.rsr' ENERGY DYNAMIC_SPHERE = on SWITCH_TRACE TRACE_OUTPUT = 1, FILE = 'topmodel6.trc' OPTIMIZE OPTIMIZATION_METHOD = 1, MAX_ITERATIONS = 20 OPTIMIZE OPTIMIZATION_METHOD = 3, TEMPERATURE = 300, MAX_ITERATIONS = 50 OPTIMIZE OPTIMIZATION_METHOD = 1, MAX_ITERATIONS = 20 ENERGY WRITE_MODEL FILE = 'topmodel6.B' But it is not including heteroatoms in the energyminimized model.But the initial model has heteroatoms.How to include heteroatoms in the file. And also the module i am using is correct or i have to do any changes.Because more energy minimization also causes misthreading of the molecule.Kindly give me some suggestions.Expecting for your reply very eagerly. Yours friendly, Senthil kumar.R

2 1

[Fwd: ALIGN2D problem]
by Dan Thomas Major 14 Nov '01

14 Nov '01

-- Dan Thomas Major (at Dr. B. Fischer's lab) Bar-Ilan University Ramat-Gan, Israel

2 1