Incompatibility between pdb generated by fd_helix and mutate_bases

[Pedrocas81]

Hello,

I am trying to generate mutated pdb files to be used for simulation in Amber.
I have used the 'fiber' and 'mutate_bases' options of x3dna, and that worked very well. However, the final pdb file does not run in Amber.
Got several messages like the ones shown below.


Creating new UNIT for residue:  sequence: 32
Created a new atom named:  within residue: .R< 32>



However, if I try building a pdb with the fd_helix routine from NAB, it works perfectly with Amber, but the coordinates and atom orders are completely different from the one of x3DNA's fiber. Morever, I am unable to use 'mutate_bases' in the pdb file obtained from NAB. So I am in a conundrum.

It seems that amber somehow does not like the atom order given by x3DNA. Could you please help me?

I am attaching the pdb file generated by NAB, for which I cannot run mutate_bases (after adding the A/B strand column of course).

Thank you very much
Pedro

[xiangjun]

Hi Pedro,

Thanks for using (web) 3DNA and for posting your questions on the Forum.

I guess the issue you met is related to PDB format. By default, 3DNA fiber and x3dna_mutate generate PDB files that follow PDBV3, which use DA, DG etc for DNA nucleotides, and OP1/OP2 instead O1P/O2P for atom names.

To verify, could you please try PDB id 355d from RCSB (or PDBe) and see if it is recognized by AMBER?

Please attached a PDB file from NAB that does work with AMBER. Since both NAB and AMBER are maintained by the same group, it is not a surprise they work happily together.

Best regards,

Xiang-Jun