fix altloc to coordinate sets

[jamesmkrieger]

Old behaviour:

$ prody select "name CA" 6flr.pdb
@> WARNING failed to parse altloc 'B' at line 1821, could not read coordinates
@> WARNING failed to parse altloc 'B' at line 1829, could not read coordinates
@> WARNING failed to parse altloc 'B' at line 1833, could not read coordinates
@> WARNING failed to parse altloc 'B' at line 1837, could not read coordinates
@> WARNING failed to parse altloc 'B' at line 1841, could not read coordinates
@> WARNING failed to parse altloc 'B' at line 1845, could not read coordinates
@> WARNING failed to parse altloc 'B' at line 10264, could not read coordinates
@> WARNING failed to parse altloc 'B' at line 10272, could not read coordinates
@> WARNING failed to parse altloc 'B' at line 10276, could not read coordinates
@> WARNING failed to parse altloc 'B' at line 10280, could not read coordinates
@> WARNING failed to parse altloc 'B' at line 10284, could not read coordinates
@> WARNING failed to parse altloc 'B' at line 10288, could not read coordinates
@> WARNING failed to parse altloc 'B' at line 10292, could not read coordinates
@> 13 out of 26 altloc 'B' lines were parsed.
@> Altloc 'B' is appended as a coordinate set to atomgroup 6flr.
@> 6073 atoms and 2 coordinate set(s) were parsed in 0.06s.
@> Secondary structures were assigned to 484 residues.
@> Selection 'name CA' matched 741 atoms.
@> Selection is written into: 6flr_selected.pdb

Fixed behaviour:

$ prody select "name CA" 6flr.pdb
@> 6086 atoms and 1 coordinate set(s) were parsed in 0.05s.
@> Secondary structures were assigned to 484 residues.
@> Selection 'name CA' matched 743 atoms.
@> Selection is written into: 6flr_selected.pdb

[jamesmkrieger]

There's also a pdb writing issue associated with this.

Old behaviour (ATOM line takes next ANISOU value):

ATOM    602  CA AHIS B 234      14.092 -10.803  78.631  0.50 76.86           C  
ANISOU  602  CA AHIS B 234    13327   8826   7048    444   3365   1618       C  
.
.
.
ENDMDL                                                                          
MODEL        2
.
.
.
ATOM    602  CA  HIS B 234    13306.000 -10.729  78.621  0.50 76.86           C  
ANISOU  602  CA  HIS B 234    13327   8826   7048    444   3365   1618       C  

Fixed behaviour:

ATOM    603  CA AHIS B 234      14.092 -10.803  78.631  0.50 76.86           C  
ANISOU  603  CA AHIS B 234    13327   8826   7048    444   3365   1618       C  
ATOM    604  CA BHIS B 234      13.993 -10.729  78.621  0.50 76.83           C  
ANISOU  604  CA BHIS B 234    13306   8825   7058    438   3380   1608       C  

[jamesmkrieger]

The problem there was that we were putting ANISOU lines into the dictionary that is fed to _evalAltloc, which only knows how to handle ATOM lines. What it does is makes another coordinate set with coords from another altloc if it exists.

New fixed result:

Python 3.9.18 (main, Sep 11 2023, 13:41:44) 
Type 'copyright', 'credits' or 'license' for more information
IPython 8.15.0 -- An enhanced Interactive Python. Type '?' for help.

In [1]: from prody import *

In [2]: ag = parsePDB("6flr.pdb", altloc=None)
@> 13 out of 13 altloc 'B' lines were parsed.
@> Altloc 'B' is appended as a coordinate set to atomgroup 6flr.
@> 6073 atoms and 2 coordinate set(s) were parsed in 0.08s.
@> Secondary structures were assigned to 484 residues.

In [3]: ag
Out[3]: <AtomGroup: 6flr (6073 atoms; active #0 of 2 coordsets)>

In [4]: hisB234 = ag.select('resname HIS and chain B and resnum 234')

In [5]: hisB234
Out[5]: <Selection: 'resname HIS and... and resnum 234' from 6flr (10 atoms; active #0 of 2 coordsets)>

In [6]: [(atom.getAltloc(), atom.getName(), atom.getCoords(), atom.getAnisou(), atom.getBeta()) for atom in hisB234.ca]
Out[6]: 
[('A',
  'CA',
  array([ 14.092, -10.803,  78.631]),
  array([1.3306, 0.8825, 0.7058, 0.0438, 0.338 , 0.1608]),
  76.86)]

In [7]: hisB234.setACSIndex(1)

In [8]: [(atom.getAltloc(), atom.getName(), atom.getCoords(), atom.getAnisou(), atom.getBeta()) for atom in hisB234.ca]
Out[8]: 
[('A',
  'CA',
  array([ 13.993, -10.729,  78.621]),
  array([1.3306, 0.8825, 0.7058, 0.0438, 0.338 , 0.1608]),
  76.86)]

In [9]: writePDB('6flr_his_B_234_ca.pdb', hisB234.ca)
Out[9]: '6flr_his_B_234_ca.pdb'

New fixed output:

REMARK Selection '(ca) and (resna...and resnum 234)'
HELIX   21 AC3 HIS B  234  HIS B  234  1                                   1    
MODEL        1
ATOM      1  CA AHIS B 234      14.092 -10.803  78.631  0.50 76.86           C  
ANISOU    1  CA AHIS B 234    13306   8825   7058    438   3380   1608       C  
ENDMDL                                                                          
MODEL        2
ATOM      1  CA  HIS B 234      13.993 -10.729  78.621  0.50 76.86           C  
ANISOU    1  CA  HIS B 234    13306   8825   7058    438   3380   1608       C  
ENDMDL                                                                          
END                                                                             

[jamesmkrieger]

We actually need to overhaul the anisou handling to give each coordset a different set of them.

[jamesmkrieger]

This will also need to be included in mmcif and mmtf parsers

[jamesmkrieger]

This will also need to be included in mmcif and mmtf parsers

These don't use _evalAltlocs anyway so probably that's something for another time

[jamesmkrieger]

We'll still need tests for this too