MODELLER thinks HSEs are ATOM instead of HETATM
Hi Modellers,
I ran into a problem with PDB file 1j9n, which contains HSE residues. Each HSE line starts with "HETATM". However, when I read them into MODELLER, MODELLER thinks HSEs are ATOMs.
env = environ() env.io.hydrogen = False env.io.hetatm = True mdl = model(env) mdl.read(file = '1j9n.pdb') for chain in mdl.chains: for residue in chain.residues: if residue.name=='HSE': print residue.name, residue.hetatm ... <Residue 16:C (type HSE)> False ...
Does anyone know what is happening here? Where can I find MODELLER's lists/libraries of HETATM/ATOM residue types?
Thank you.
Yingjie - Yingjie
On 9/1/11 8:12 PM, Yingjie Lin wrote: > I ran into a problem with PDB file 1j9n, which contains HSE residues. > Each HSE line starts with "HETATM". However, when I read them into > MODELLER, > MODELLER thinks HSEs are ATOMs.
Modeller does not read the ATOM/HETATM information from PDB - it only uses it to decide which residues to read in (those marked as HETATM are only read in if you set env.io.hetatm=True).
> Where can I find MODELLER's lists/libraries of HETATM/ATOM residue types?
See modlib/restyp.lib. The first column in this file tells you whether Modeller treats a given residue type as ATOM or HETATM.
Ben Webb, Modeller Caretaker
On 9/1/11 8:12 PM, Yingjie Lin wrote: > I ran into a problem with PDB file 1j9n, which contains HSE residues.
I should also note that Modeller names HSE and HSP using CHARMM convention, not PDB, in which these are used to refer to different protonation states of histidine. Since PDB since remediation now uses HSE and HSP for certain ligands (not histidine) this will give the odd results you're seeing when you encounter HSE/HSP residues. It should be sufficient to edit modlib/restyp.lib and remove the two lines for HSE and HSP to make the problem go away. I will do this for the next Modeller release.
Ben Webb, Modeller Caretaker
participants (2)
-
Modeller Caretaker -
Yingjie Lin