5-Bromo-2′-deoxyuridine monomer in coot and Jligand

From: zhang yu
Date: 2011/10/23


Hi,

My DNA chain has a internal modification of 5-Bromo-2′-deoxyuridine (5BrdU). I had a problem when I tried to model 5BrdU into my DNA chain by coot. I will appreciate it very much if someone can solve it for me.

What I did...
1. Generate the cif dictionary and export the pdb for 5BrdU in "Jligand".
2. Load the pdb and cif into coot
4. Merge the monomer into my molecule. Change and renumber 5BrdU (Chain G and resi 15).
3. Run "Real-space refine" in coot

The problem...
"Real-space refine" fits the monomer into density quite well, but coot only makes bond (P-O) between 5BrdU and its upstream nt (14), doesn't make bonds between 5BrdU and its downstream nt (16). Whenever I refine it, it seems coot just treats 5BrdU and nt 16 as in different chain, push two atoms away(P in nt16 and O3 in nt15 ),  and doesn't connect them.

Did I miss something when I generate cif dictionary or at other steps?

Yu
   

--
Yu Zhang
HHMI associate
Waksman Institute, Rutgers University
190 Frelinghuysen Rd.
Piscataway, NJ, 08904



----------
From: Sabine Schneider

Hi Yu,

A few things which caused this happened to me with other non-standart
nucleotides:

- the hydrogen of the 3'O of the BrdU is defined as 3'OH in the cif file
- order of the atoms in the pdb file ie compare the position of the 3'OH
with the standard dNTP in the PDB
- is the 3BrdU defined as 'DNA' in the cif file? (under 'data_comp_list')

Maybe that helps!

Sabine

----------
From: zhang yu

Hi Sabine,

Thanks for your suggestions.

I checked the cif file of BrdU.

the hydrogen of the 3'O of the BrdU is defined as HO3' in BrdU cif file, in stead of 3'OH. The standard dNTP cifs define it as HO3' too.
In PDB, there is no atoms for hydrogen, as I excluded them.
The BrdU is defined as "UBP" as I named in my cif file (under 'data_comp_list')

Any suggestion? I attached my cif and pdb of BrdU, could you take a look at it?
Do you have any suggestion of program generating cif for non-standard nucleotides?

Thanks

Yu

在 2011年10月23日 下午3:50，Sabine Schneider 

----------
From: Ed Pozharski


As Sabine said, you need to make sure that the monomer is defined as DNA
type.  The best way to figure out how to do this is to look at the
standard monomer library for nucleotide, e.g.
$CCP4_LIB/data/monomers/d/DA.cif.

--
Oh, suddenly throwing a giraffe into a volcano to make water is crazy?
                                               Julian, King of Lemurs

----------
From: Alan Cheung


Hi Yu - we regularly model BrdU in our DNA chains with the "replace
residue" function in coot.  Model a normal thymine in the required
position and then centre on the thymine.  Then select :

Extensions > Modelling > Replace Fragment

and type BRU into the dialog box.

(see 5.17.3 of
http://biop.ox.ac.uk/coot/doc/coot/Mutation.html#Mutation )

It will also add a superfluous OP3 atom to the BRU phosphate atom upon
residue replacement which you'll need to delete.

Make sure the flanking nucleotides of DNA can already be
regularised/refined with coot (a version dependent issue i think)
otherwise it won't link the BRU properly.

Alan

----------
From: Alan Cheung

Whoops, i meant

Extensions>  Modelling>  Replace Residue

and not replace fragment.

----------
From: zhang yu

Thanks, I appreciate all the help from you guys. 

A brief summary,

As Sabine, Paul and Ed pointed, change the "_chem_comp.group" in "data_comp_list" to DNA is the key for coot to recognize it.

As Alan said, "Extension>Modeling>Replace Residues" could also insert BRU into DNA chain without any modification of cif file.