Dear Ibrahim,
> I have loaded both leaprc.ff12SB and leaprc.GLYCAM_06h for my dipeptide
> (Lys-Fru) as non-standard amino acid residue. Then I have loaded the new
> FF.lib with the head and tail were defined and the pdb file of the protein
> in xleap but I still have a problem with atom type. I got the following
> error
> Unit Editor: > check x
> Unit Editor: Checking 'x'....
> Unit Editor: WARNING: The unperturbed charge of the unit: -1.000000 is not
> zero.
oh oh... Why the total of 'x' is not neutral? Lys-Fru should be
neutral & perfectly = 0.0000 if you do use R.E.D. So my guess is that
'x' is the whole molecule that contains this Lys-Fru residue.
I do:
lynx:~> tleap
-I: Adding /usr/local/amber12/dat/leap/prep to search path.
-I: Adding /usr/local/amber12/dat/leap/lib to search path.
-I: Adding /usr/local/amber12/dat/leap/parm to search path.
-I: Adding /usr/local/amber12/dat/leap/cmd to search path.
Welcome to LEaP!
(no leaprc in search path)
> loadoff FLY.off
desc FLY
UNIT name: FLY
Head atom: .R<FLY 1>.A<N 1>
Tail atom: .R<FLY 1>.A<C 41>
Contents:
R<FLY 1>
> desc FLY.1
RESIDUE name: FLY
RESIDUE sequence number: 1
RESIDUE PDB sequence number: 0
Type: saccharide
Connection atoms:
Connect atom 0: A<N 1>
Connect atom 1: A<C 41>
[...]
--> looks ok
> charge FLY
Total unperturbed charge: 0.000000
Total perturbed charge: 0.000000
--> looks ok
> savemol3 FLY FLY.mol2 1
Writing mol3 file: FLY.mol2
-> We can display the FF lib in VMD: this looks pretty good...
> Unit Editor: Warning: Close contact of 0.930843 angstroms between .R<PHE
> 11>.A<H 2> and .R<SER 34>.A<HG 9>
> Unit Editor: Warning: Close contact of 1.107358 angstroms between .R<TYR
> 50>.A<H 2> and .R<SER 77>.A<HG 9>
> Unit Editor: Warning: Close contact of 1.198180 angstroms between .R<PHE
> 56>.A<H 2> and .R<GLN 143>.A<HE21 14>
> Unit Editor: Warning: Close contact of 1.323347 angstroms between .R<ARG
> 59>.A<HH11 18> and .R<ARG 59>.A<HD2 12>
> Unit Editor: Warning: Close contact of 1.004213 angstroms between .R<PHE
> 98>.A<H 2> and .R<SER 123>.A<HG 9>
> Unit Editor: Warning: Close contact of 1.432209 angstroms between .R<ARG
> 99>.A<HE 15> and .R<ARG 99>.A<HB2 6>
> Unit Editor: Warning: Close contact of 1.378498 angstroms between .R<TYR
> 134>.A<HH 15> and .R<ARG 168>.A<HE 15>
> Unit Editor: Warning: Close contact of 0.982326 angstroms between .R<TYR
> 139>.A<H 2> and .R<SER 166>.A<HG 9>
you have close contacts in your structure; they are is only warnings,
but you should look at them and see if this is expected or not...
> Unit Editor: Checking parameters for unitLys-Fru 'x'.
> Unit Editor: Checking for bond parameters.
> Unit Editor: Checking for angle parameters.
> Unit Editor: Could not find angle parameter: CT - NT - Cg
> Unit Editor: There are missing parameters.
> Unit Editor: check: Warnings: 9
> Unit Editor: Unit is OK.
>> quit
> Unfotunately I can not find this angle parameter. Please, any suggestions
> will be appreciated. Please I attached both my dipeptide (FLY.off) and a
> pdb file for protein (protein.pdb).
Here you mix FF and even atom types; so I am not surprised you get
missing FF parameters. This is why at
http://q4md-forcefieldtools.org/REDDB/projects/F-85/ a single 'type'
of FF 'atom types' is used...
CT-NT-Cg is just CT-NT-CT (my guess); just create a frcmod file and
duplicate this FF parameters, but here you clearly see that loading
two force fields is far more complex than using a single one; CT-NT-CT
is already known in parm99.dat:
grep "CT-NT-CT" parm99.dat
CT-NT-CT 50.0 109.50 neutral amines
In your case you have loaded two FF (or may be three FF if you load
gaff with different atom types) where FF parameters/atom types are
renamed/duplicated/copied/... My feeling is that this is far too
complex considering that Lys-Fru is unlikely to be already
parameterized by the last version of Glycam 20006: just my personal
opinion...
regards, Francois
> On Fri, Feb 22, 2013 at 9:17 AM, FyD <fyd.q4md-forcefieldtools.org> wrote:
>
>> Dear Ibrahim,
>>
>> > I Still have one more question (probably). I am trying to link my
>> dipeptide
>> > (LYS-FRU fragment) back into my protein. In the protein, LYS residue is
>> in
>> > the second position and when load my pdb file into xleap an extra OXT
>> atom
>> > automatically added to the first residue and two hydrogen atoms to the
>> the
>> > third residue.
>>
>> OXT means C-terminal and consequently means last residue; not first one...
>>
>> If OXT is automatically added by LEaP this means it is missing in the
>> C-terminal fragment in your PDB file: you have to check that...
>>
>> More generally once LEaP has identified a residue (i.e. match between
>> a FF lib & a residue from a PDB file) LEaP automatically adds the
>> missing atoms (missing in the PDB file if any) based on these defined
>> in the FF lib.
>>
>> > I am using the following script for LYS-FRU Fragment:
>> > ./xleap -s -f leaprc.ff12SB
>> > glycam_06 = loadamberparams GLYCAM_06h.dat
>> > X = loadmol2 X.mol2
>> > set X head X.1.N
>> > set X tail X.1.C16
>> > set X.1 connect0 X.1.N
>> > set X.1 connect1 X.1.C16
>>
>> -> I would define atom names in the new FF library that means
>> 'something'; i.e. CA for alpha carbon by analogy to the other residues
>> available in the Amber force field topology database (& in particular
>> with the regular LYS residue).
>>
>> This should allow you to more easily identify atoms when displaying a
>> structure in a graphical program and when using ptraj you will be able
>> to use * and ? cards for instance all the CA atoms...
>>
>> > set X.1 restype saccharide
>> > set X.1 name "mol"
>> > set X.1.C1 type CG
>> > .
>> > set X.1.H62 type H1
>> > then I loaded my protein into xleap. Now, How can I bind the LYS-FRU to
>> the
>> > 2nd and 3rd residues (the preceding and the following residues).
>>
>> Jason has just answered to this question in an another email:
>> Let's count this LYS-FRU residue as the i residue: once the head/tail
>> are defined in the FF lib the connection between the pairs of residues
>> "i-1 - i" and "i - i+1" are automatic.
>>
>> - You load all the Amber FF lib in LEaP
>> - You load your new FF lib with head/tail defined
>> - You load the PDB file (exp. data)
>> - Residues are recognized if the atom & residue names are identical in
>> the PDB file & FF lib
>> - Connections between pairs of residues is carried out if the
>> head/tail are defined for all the residues
>>
>> regards, Francois
>>
>>
>> > On Sun, Feb 17, 2013 at 2:38 PM, FyD <fyd.q4md-forcefieldtools.org>
>> wrote:
>> >
>> >> Dear Ibrahim,
>> >>
>> >> > I got the following file for capped Lys-Frupyr
>> >> > fragment. The RED server name P2909, but I recognized some files
>> active
>> >> and
>> >> > others are inactive. Also, I don' t know if these large number of
>> files
>> >> > indicates the QM run is Ok.
>> >>
>> >> You ran a big job and automatically generated FF libraries for the
>> >> N-term, C-term & central fragments + the corresponding dipeptide
>> >> according to the tutorial at:
>> >> http://q4md-forcefieldtools.org/Tutorial/Tutorial-3.php#25
>> >>
>> >> Yes, a lot of files were generated:
>> >> http://cluster.q4md-forcefieldtools.org/~.../Project/P2909.html
>> >> this is normal you get lost ;-)
>> >>
>> >> All the files generated by R.E.D. Server are described at:
>> >>
>> http://q4md-forcefieldtools.org/Tutorial/P2N/All-frag-Pept/listing-6mol.pdf
>> >> see page 5: you really need:
>> >> mm1-o1.FG2.mol2 Central fragment: Molecule 1 - conformation 1 (Fragment
>> 2)
>> >> mm3-o1-FG.mol2 N-terminal fragment: Molecule 3 - conformation 1
>> >> mm5-o1-FG.mol2 C-terminal fragment: Molecule 5 - conformation 1
>> >> mm6-o1.mol2 Molecule 6 - conformation 1
>> >> & likely only:
>> >> mm1-o1.FG2.mol2 Central fragment: Molecule 1 - conformation 1 (Fragment
>> 2)
>> >>
>> >> The key P2N file is the first one of the six files:
>> >> http://cluster.q4md-forcefieldtools.org/~.../Project/P2909/Mol_red1.p2n
>> >> -> chemical equivalencing was generated by Ante_R.E.D. 2.0 -> all is
>> ok:
>> >> http://q4md-forcefieldtools.org/REDS/news.php#2
>> >>
>> >> REMARK Information automatically added by R.E.D. Server
>> >> REMARK INTRA-MCC 0.0000 | 1 2 3 4 5 6 | Remove
>> >> REMARK INTRA-MCC 0.0000 | 49 50 51 52 53 54 | Remove
>> >> -> this is to define the central fragment
>> >>
>> >> REMARK INTER-MCC 0.0000 | 2 3 | 1 2 3 4 | 1 2 3 4 5 6 7 8
>> >> REMARK INTER-MCC 0.0000 | 4 5 | 1 2 3 4 | 47 48 49 50 51 52 53 54
>> >> -> this is to define the N-term & C-term fragments between molecules
>> >> 2-3 & 4-5
>> >>
>> >> the RRMS of the RESP fit with chemical equivalencing and all the
>> >> charge constraints:
>> >>
>> >>
>> http://cluster.q4md-forcefieldtools.org/~.../Project/P2909/Data-R.E.D.Server/Mol_MM/output2_mm
>> >>
>> >> ESP relative RMS (SQRT(chipot/ssvpot)) 0.03331
>> >>
>> >> the RRMS of the 1st RESP stage without chemical equivalencing and
>> >> without the charge constraints:
>> >>
>> >>
>> http://cluster.q4md-forcefieldtools.org/~.../Project/P2909/Data-R.E.D.Server/Mol_MM/stat-output-b_mm
>> >>
>> >> ESP relative RMS (SQRT(chipot/ssvpot)) 0.02964
>> >>
>> >> -> the difference 0.033-0.030 is pretty good...
>> >>
>> >> You can also look at local/charge errors:
>> >>
>> >>
>> http://cluster.q4md-forcefieldtools.org/~.../Project/P2909/Data-R.E.D.Server/Mol_MM/stat-RESP-FIT_mm
>> >> -> the errors are mainly observed for N-term & C-term fragments; this
>> >> is normal...
>> >>
>> >> You decided to use 'pop=mk' for all your data; I think this is the
>> >> most simple; at least in a first approach:
>> >>
>> >>
>> http://cluster.q4md-forcefieldtools.org/~.../Project/P2909/Data-R.E.D.Server/Mol_m1/JOB2-gau_m1-1-1.com
>> >>
>> >> The key to finish to check your RESP charge derivation data:
>> >> do you like the conformation obtained after geometry optimization?:
>> >> See the java applet :
>> >>
>> >>
>> http://cluster.q4md-forcefieldtools.org/~.../Project/P2909/javaappletpdb-1.html
>> >> The peptide backbone looks like to be in an extended conformation;
>> >> concerning the fructose part (http://en.wikipedia.org/wiki/Fructose),
>> >> the conformation is not 4C1
>> >> (http://web.inc.bme.hu/csonka/csg/glc/glc.htm); only you should know
>> >> what conformation you want for this fructose part: it should be close
>> >> to your experimental data or close to a choice you voluntary made!
>> >>
>> >> I hope this helps...
>> >>
>> >> regards, Francois
>> >>
>> >>
>> >> > On Fri, Feb 15, 2013 at 6:09 PM, FyD <fyd.q4md-forcefieldtools.org>
>> >> wrote:
>> >> >
>> >> >> Ibrahim,
>> >> >>
>> >> >> I continue my former email... In case you are interested in the mol3
>> >> >> file format you could have a look at the 'F-93' R.E.DD.B. project by
>> >> >> J. Sanders about peptide nucleic acid; this is an example where the
>> >> >> mol3 file format is used... This simplifies I think a lot the use of
>> >> >> force field libraries and the connections between fragments in the
>> >> >> LEaP program...
>> >> >> See http://q4md-forcefieldtools.org/REDDB/projects/F-93/
>> >> >>
>> >> >> regards, Francois
>> >> >>
>> >> >>
>> >> >> > thank you for helping. My R.E.D server name P2890 for Lys-Frupyr
>> file.
>> >> >> Now
>> >> >> > I have the following problems: (1) how can ligate this dipeptide
>> into
>> >> my
>> >> >> > protein. of course, I need to assign atom types, (2) Can I use
>> xleap
>> >> and
>> >> >> > then add the fructose molecule to lysine residue in my protein
>> >> manually,
>> >> >> > then select this residue and assign both atom types and add charges
>> >> from
>> >> >> > P2890 file manually. The problem of later is, if I dismessed, I
>> need
>> >> to
>> >> >> > re-add atom types and charges in xleap. Please, any suggestions.
>> >> >>
>> >> >> > On Thu, Feb 14, 2013 at 9:00 AM, FyD <fyd.q4md-forcefieldtools.org
>> >
>> >> >> wrote:
>> >> >> >
>> >> >> >> Dear Ibrahim,
>> >> >> >>
>> >> >> >> > I have built my dipeptide (Lysine-fructose). and I used RESP ESP
>> >> >> charge
>> >> >> >> > derived server to assign the atomic charges. Please, I used
>> xleap
>> >> to
>> >> >> add
>> >> >> >> > the same fructose moiety to the lysine residue in my protein and
>> >> added
>> >> >> >> the
>> >> >> >> > charges and atom types to this fructose part manually. Of
>> course, I
>> >> >> >> loaded
>> >> >> >> > both of leaprc.ff99SB and Glycam_06h. Now I get both top and crd
>> >> files
>> >> >> >> but
>> >> >> >> > I do not know if these steps are Ok or no. please can you help
>> me.
>> >> >> >>
>> >> >> >> This is difficult to help with so little information...
>> >> >> >>
>> >> >> >> -> I guess you generated a dipeptide for this Lysine-fructose and
>> >> used
>> >> >> >> R.E.D. Server to generate a central fragment for this modified
>> >> >> >> dipeptide. May be you could provide in your email the 'PXXXX'
>> R.E.D.
>> >> >> >> Server name so that we can more easily assist you by looking at
>> your
>> >> >> >> R.E.D. Server job.
>> >> >> >>
>> >> >> >> To generate a central fragment for your modified dipeptide see:
>> >> >> >> http://q4md-forcefieldtools.org/Tutorial/Tutorial-3.php#15
>> >> >> >>
>> >> >> >> To generate the N-term, C-term + central fragments manually, see:
>> >> >> >> http://q4md-forcefieldtools.org/Tutorial/Tutorial-3.php#24
>> >> >> >>
>> >> >> >> To generate these 3 fragments automatically see:
>> >> >> >> http://q4md-forcefieldtools.org/Tutorial/Tutorial-3.php#25
>> >> >> >>
>> >> >> >> key points here:
>> >> >> >> - is chemical equivalencing correctly defined in the P2N input
>> file?
>> >> >> >> See http://q4md-forcefieldtools.org/REDS/news.php#2
>> >> >> >>
>> >> >> >> - what is/are the conformation(s) involved in charge derivation
>> for
>> >> >> >> your dipeptide with this Lysine-fructose?
>> >> >> >>
>> >> >> >> - what is the algorithm used in MEP computation considering that
>> you
>> >> >> >> use two different force fields which are based on different MEP
>> >> >> >> computation algorithm (Connolly surface vs CHELPG). To simplify
>> all
>> >> >> >> that for this fructose-based central fragment of Lys, you might be
>> >> >> >> interested in an approach we develop for a-1,4 Glc oligomers:
>> >> >> >> see http://www.ncbi.nlm.nih.gov/pubmed/21792425 : the main idea
>> in
>> >> >> >> this paper is to provide a highly consistent approach for force
>> field
>> >> >> >> development for glycopeptides; however only tested for a-1,4 Glc
>> >> based
>> >> >> >> oligomers. this means that if you decide to follow this approach
>> you
>> >> >> >> will have to validate it (in all the cases you always have to
>> >> validate
>> >> >> >> your approach; so no difference...)
>> >> >> >>
>> >> >> >> -> then at the end R.E.D. Server/R.E.D. IV provides a mol2 file
>> or a
>> >> >> >> series of mol2 files that you have to load in LEaP. Here you need
>> to
>> >> >> >> decide which force field(s) you plan to use and define the
>> >> >> >> corresponding atom types. Personally I always add these atom types
>> >> >> >> manually because I want to control my choices.
>> >> >> >>
>> >> >> >> -> finally you load all the FF libs, define the head/tails (to
>> >> connect
>> >> >> >> them where they should be connected) in the LEaP program and
>> generate
>> >> >> >> the prmtop/prmcrd files; if force field parameters are missing
>> LEaP
>> >> >> >> will generate errors/the listing of these missing parameters; this
>> >> >> >> means you have to generate a frcmod file; once again I always
>> >> generate
>> >> >> >> this frcmod file by hand to control my choices. See for instance:
>> >> >> >>
>> >> >> >> the R.E.DD.B. project in relation with the publication above
>> >> >> >> http://q4md-forcefieldtools.org/REDDB/projects/F-85/
>> >> >> >>
>> >> >> >> the definition of head/tail and FF atom types
>> >> >> >> http://q4md-forcefieldtools.org/REDDB/projects/F-85/script1.ff
>> >> >> >>
>> >> >> >> the frcmod file
>> >> >> >> http://q4md-forcefieldtools.org/REDDB/projects/F-85/script3.ff
>> >> >> >>
>> >> >> >> I hope this helps...
>> >> >> >>
>> >> >> >> regards, Francois
>> >> >> >>
>> >> >> >>
>> >> >> >> > On Thu, Jan 24, 2013 at 9:51 AM, FyD <
>> fyd.q4md-forcefieldtools.org
>> >> >
>> >> >> >> wrote:
>> >> >> >> >
>> >> >> >> >> Dear Ibrahim,
>> >> >> >> >>
>> >> >> >> >> > I am trying to simulate the Schiff base for my protein in
>> which
>> >> >> >> >> > lysine is binding the fructose molecule through a covalent
>> bond
>> >> >> NZ-C1
>> >> >> >> of
>> >> >> >> >> > fructose. Can I use xleap to build this structure and how.
>> What
>> >> I
>> >> >> >> have in
>> >> >> >> >> > my hand that I should separate lysine residue from the
>> protein
>> >> and
>> >> >> >> attach
>> >> >> >> >> > it to fructose and then re-ligate, is it correct. I do not
>> know
>> >> how
>> >> >> >> can
>> >> >> >> >> > xleap can do.
>> >> >> >> >>
>> >> >> >> >> If I understand you you are interested in constructing a new
>> >> residue
>> >> >> >> >> i.e. a L-lysine connected to D-fructose through an imine bond.
>> >> >> >> >>
>> >> >> >> >> You can use R.E.D. and/or R.E.D. Server for this work; and you
>> can
>> >> >> >> >> directly build a new central fragment for this new Lysine
>> residue.
>> >> >> >> >> See for instance:
>> >> >> >> http://q4md-forcefieldtools.org/Tutorial/Tutorial-3.php
>> >> >> >> >> &
>> >> >> >> >> http://q4md-forcefieldtools.org/Tutorial/Tutorial-3.php#24
>> >> >> >> >> vs
>> >> >> >> >> http://q4md-forcefieldtools.org/Tutorial/Tutorial-3.php#25
>> >> >> >> >>
>> >> >> >> >> You could also consider splitting this 'lys-fructose' residue
>> into
>> >> >> two
>> >> >> >> >> building blocks. You can find examples of such an approach in
>> >> >> R.E.DD.B.
>> >> >> >> >> See for instance in the sugar domain:
>> >> >> >> >> http://q4md-forcefieldtools.org/REDDB/projects/F-85/
>> >> >> >> >> http://q4md-forcefieldtools.org/REDDB/projects/F-72/
>> >> >> >> >> http://q4md-forcefieldtools.org/REDDB/projects/F-71/
>> >> >> >> >>
>> >> >> >> >> R.E.D. uses the P2N file format as input described at:
>> >> >> >> >> http://q4md-forcefieldtools.org/Tutorial/Tutorial-1.php#3
>> >> >> >> >> & generates FF library in the mol2 file format described at:
>> >> >> >> >> http://q4md-forcefieldtools.org/Tutorial/leap-mol2.php
>> >> >> >> >> This mol2 file format is directly usable in the LEaP prgram as
>> >> >> described
>> >> >> >> >> at:
>> >> >> >> >> http://q4md-forcefieldtools.org/Tutorial/Tutorial-1.php#1
>>
_______________________________________________
AMBER mailing list
AMBER.ambermd.org
http://lists.ambermd.org/mailman/listinfo/amber
Received on Wed Feb 27 2013 - 02:00:03 PST
