1JTW error

[Diego Gallego]

Hi! I recently discovered that the PDB 1JTW gives an error when I try to read multiple models:

library(bio3d); pdb <- read.cif("1JTW", multi=T)

Note: Accessing on-line CIF file

Warning messages: 1: In read.cif("1JTW", multi = T) : helix/sheet records will not be parsed in this version of the code

2: In get.pdb(file, path = tempdir(), quiet = TRUE, format = "cif") : /tmp/Rtmp6NsTxE/1JTW.cif exists. Skipping download

3: In matrix(pdb$xyz, nrow = pdb$models, byrow = TRUE) : data length [22536] is not a sub-multiple or multiple of the number of rows [14]

4: In as.xyz(pdb$xyz) : number of cartesian coordinates not a multiple of 3

I though it was worth sharing with you! Cheers,

Diego

[Lars Skjærven]

Thanks for reporting Diego. There are unequal number of atoms between the different models in the PDB file. i.e 537, 538, 539 and 524 atoms.

The previous version of read.pdb , (found in function read.pdb2()) provides a cleaner error message:

> pdb = read.pdb2("1jtw.pdb", multi=T)
  PDB has multiple END/ENDMDL records 
  multi=TRUE: 'read.dcd/read.ncdf' will be quicker! 
  HEADER    RNA                                     22-AUG-01   1JTW               
Warning message:
In read.pdb2("1jtw.pdb", multi = T) :
  Unequal number of atoms in multi-model records: 1jtw.pdb

Todo:

• cleaner error messages for read.cif and new read.pdb
• explore ways to account for the different number of atoms

[Xinqiu Yao]

Not a bug but because of bad format of input files.