Re: [AMBER] Question about xleap: How to define a linkage between DNA and phosphotyrosine.

From: Catein Catherine <>
Date: Mon, 5 Jul 2010 17:39:58 +0800

Dear Francois,


Thank you very much for your feedback. I agree with you that I need to build a new model. Before i start, I could like to ask for your additional advices on these issues. Many thanks in advance for your help.


(1) It is no doubt that i need to build a PTR residue. If I understand your tutorials correctly, I should add ACE cap to the N-terminal of PO2-Tyr residue, and add NME cap to the C-terminal of PTR. (while both ACE and NME should have zero charge for the whole procedures). (Note: only two Oxygen attached to P)


(2) Then, I have to add an OMe cap to the P atom of PTR residue, again I should assign OMe has total charge of zero.


(3) Do the optimization and ESP for the this PTR residues, by keeping the total charge of the whole residue = -2.


(4) I check the head and tail of CXY by typing "desc CYX" in xleap. I found Head atom: R<CYX1>. A<N 1> Tail atom: .R<CYX 1>. A(C9>. Where can found the additional head or tail for CYX? Could you mind to let me know the better procedure to find the head and tail atoms of a unit? I also wonder the command should be used to add two tails for my PTR residue.


By doing these, I think should have finished PTR residue setting.


For the DT part, I have two thoughts.


(1) keep DT and its next steps (another DA, not connected, as the DNA also break at this site) as internal DT/DA. However, I encounter problem when I try to load the pdb file of the whole DNA-protein complex. As DT and DA are not connected, I found xleap automatically convert them to DT3 and DA5, even I manually correct them to DA and DT and removed the ter card in the pdb file. I don't want to use the command line of clearpdbresmap, as I want xleap to correct the geometries or connections for the rest of pdb file. Could you mind to give me some hints how to do it?


(2) define DT as DT3, and let the xleap to automatically change DA to DA5. By doing this, I may have to design a new residue, like DTx, which have a total charge of -0.692100. However, i will encounter two problems here. (i) I need a cap to have a charge of -0.307900 so as to run RED.III.x. (iii) Then, I have to assign the O3' as the tail for DTx.


Which way is more reasonable?


Many thanks in advance,


Best regards,


> Date: Mon, 5 Jul 2010 08:44:44 +0200
> From:
> To:
> CC:
> Subject: Re: [AMBER] Question about xleap: How to define a linkage between DNA and phosphotyrosine.
> Dear Catherine,
> - I do not think using a central nucleotide fragment in place of a
> terminal one (or vice-versa) is a good idea. If you do so it is
> unlikely the total charge of your system will remain an integer (what
> you want to reach) and you will generate open valencies. For
> nucleotide fragments, you need to keep in mind that the central
> fragment takes an integer value (not always = -1; see the F-83
> R.E.DD.B. project by L. Tao for instance) and the sum of the 3' & 5'
> terminal fragments takes also an integer value.
> See also the total charge of DA, DA3 & DA5 in the Amber force field
> topology database:
> > charge DA
> Total unperturbed charge: -1.000000
> > charge DA3
> Total unperturbed charge: -0.692100
> > charge DA5
> Total unperturbed charge: -0.307900
> -0.692100-0.307900 = -1.0000 !
> - A consequence is that I do not thing you can use the phosphorylated
> Tyrosine published by Homeyer et al.: the connections between the
> nucleotide & peptide parts do not match: the resulting total charge
> will not be an integer and you will have either missing oxygens and
> "too much" ;-) oxygens in the connections.
> You need to build (a) new fragment(s) for your phosphorylated amino
> acid (PAA) in agreement with the nucleotide fragments already
> available. If I well understood your system, your PAA is located at
> the 3' terminal. But you cannot use the 3' regular terminal fragment
> which is 3'OH. Thus, you need to add a PAA fragment which has the same
> total charge as DA3, and this PAA fragment is in fact a Tyr-O-PO2
> fragment.
> Your case is quite interesting/tricky. I suggest you to use
> R.E.D.-III.x to learn how to use intra- & inter-molecular charge
> constraints during the charge fit to define molecular fragments. Then,
> you might decide to use R.E.D. IV/R.E.D. Server if your case is too
> complex.
> See
> You might find answer in more complex projects:
> additional answers are below.
> > I would like to define the linkage between a phosphotyrosine and a
> > DNA base pair. I make myself clear, I have prepare the model
> > structure as the attached pdb file. This is exactly the model that
> > I could like to define for amber calculation. I already found the
> > parameters for phosphotyrosine (I downloaded from the amber webpage
> > previously). I have difficulty to define a linkage between DT and
> > PTR. I have serveral questions would like to ask for your guidence.
> This is not because you found FF libraries for something that has the
> same name that you can apply it to your system! I do not think the PAA
> FF libraries from Homeyer et al. are compatible with what you need.
> > (1) how to add a bond between O3P of PTR and C3' of DT in xleap?
> > What is the text command should be used?
> Look at the connecting atoms for DA and the PAA of Homeyer et al.:
> they do not match to do what you need.
> > (2) there are four predefined residues for DT in ff99SB force
> > fields, i.e. DT3, DT4, DTN and DT. however, all of them has a
> > Oxygen atom attached to the C3', if I define the pdb file using
> > anyone of them. It may be problematic, as PTR will also have an
> > additional oxgyen.
> Yes so you need to build your own PAA fragment keeping in mind that
> the total charge of your new PAA fragment + DA5 (or DX5) takes an
> integer value (-1 here).
> > (3) I think I should define a new unit (name as DTp) in xleap to
> > represent a "DT without Oxygen in C3'".
> Yes
> > In order to maintance the charge of the DTp unit same as DT. I may
> > modify the charge of DTp by removing the oxygen atom from DT, and
> > share the charge of oxygen to all the rest of atoms in DTp. Do you
> > think it is reasonable approach?
> I would create (a) new PAA fragment(s)
> > (4) I also have to solve the problem head and tail problems in the
> > new units. For DTp, the head should be the other base pairs one
> > step above DTp. the tail should be PTR. However, for PTR, as it is
> > a protein residues, the head and tail should be the residue that
> > link the N-terminal or C-terminal of tyrosine. However, do I have
> > to define PTR has another tail or head at the O3P atom? If yes, how
> > to define more than one tails in xleap?
> No problem here; your PAA fragment will have three connect atoms like
> CYX in disulfide bridge. Look at CYX and you will have all what you
> need for your PAA fragment.
> > (5) I also would like to ask the text command should be used to
> > remove an atom in xleap, as it could be very difficult to pick an
> > atom in edit mode of xleap.
> This is not just deleting an atom (or adding one) you also need to get
> an integer value for the total charge of your system.
> The best here might to re-generate the terminal & central fragments...
> This is a little brain storming & this just like playing LEGO ;-) I
> hope this helps.
> regards, Francois
> _______________________________________________
> AMBER mailing list
Hotmail: Free, trusted and rich email service.
AMBER mailing list
Received on Mon Jul 05 2010 - 03:00:03 PDT
Custom Search