def write_pdb_file(self,filename,options):
"""
Writes a pdb file containing all valid atoms. The bfactor option
may be set to 'packdens' - substitutes bfactor with packing density.
'volume' - same with total atomic volume
'minuspackdens' - same with 1-packing density
'zscore' - z-score (normalized difference to reference value)
"""
bfactor = options['bfactor']
out = []
# z-score calculation only needed when the b-factor is to
# be replaced with the z-score
# actually, the expensive part is not the ZScore initialization
# but the AveragePacking initialization
if bfactor == 'zscore':
zsc = ZScore(AveragePacking(options))
for i in range(len(self.atoms)):
if not self.valid[i]:continue
atom = self.get_atom(i)
pl = atom['pdb_line']
# now get the value to be inserted as bfactor
#
# For some programs they need to be normalized
#
# Hallo Fiete!
#
# unter 'packdens' kannst Du dich austoben.
# Habe ein Beispiel reingeschrieben,
# in Python sollte etwas Mathe nicht so schwer sein.
# Aber frag mich trotzdem falls Du im Code versumpfst.
# 11.7.06 KR
#
if bfactor == 'packdens':
value = atom['packing_density']
# normalize values from 0.4 to 0.9 to bfactors between 0.0 and 30.
# 'packdens'
vmin = 0.4
vmax = 0.9
bmin = 0.0
bmax = 30.0
value = ((value-vmin)/vmax) * (bmax-bmin) + bmin
#
if bfactor == 'minuspackdens':
value = 1-atom['packing_density']
# normalize values from 0.4 to 0.9 to bfactors between 30.0 and 0.0
# 'minuspackdens'
vmin = 0.4
vmax = 0.9
bmin = 0.0
bmax = 30.0
value = ((value-vmin)/vmax) * (bmax-bmin) + bmin
#
elif bfactor == 'volume':
value = atom['total_volume']
# nomalize
# (insert code on demand)
elif bfactor == 'zscore':
value = zsc.get_z_score(atom,options)
if not value: value=100.0
# normalize
# (insert code on demand)
pl = pl[:60]+"%6.2f"%(value)+pl[66:]
out.append(pl+'\n')
open(filename,'w').writelines(out)
def replace_bfactors(self,infile,outfile,options,check_valid=False):
"""
Writes the .vol file 'infile' to 'outfile', but replaces the bfactor
column with either packing density or volume.
Also adds a short remark before the first atom.
"""
extralines_pd="""REMARK
REMARK COLUMNS MODIFIED FROM ORIGINAL PDB FILE:
REMARK PD - LOCAL PACKING DENSITY
REMARK VDWVOL - VOLUME INSIDE VAN-DER-WAALS SPHERE
REMARK SEVOL - VOLUME IN 1.4 ANGSTROM LAYER OUTSIDE VDW-SPHERE
REMARK BUR - INDICATES BURIED ATOMS (1=BURIED, 0=SURFACE)
REMARK
REMARK AT# ELEM RES C RES# X Y Z PD PD VDWVOL SEVOL BUR
"""
extralines_vol="""REMARK
REMARK COLUMNS MODIFIED FROM ORIGINAL PDB FILE:
REMARK VOL - TOTAL ATOMIC VOLUME
REMARK VDWVOL - VOLUME INSIDE VAN-DER-WAALS SPHERE
REMARK SEVOL - VOLUME IN 1.4 ANGSTROM LAYER OUTSIDE VDW-SPHERE
REMARK BUR - INDICATES BURIED ATOMS (1=BURIED, 0=SURFACE)
REMARK
REMARK AT# ELEM RES C RES# X Y Z VOL VOL VDWVOL SEVOL BUR
"""
extralines_z="""REMARK
REMARK COLUMNS MODIFIED FROM ORIGINAL PDB FILE:
REMARK Z - Z-SCORE
REMARK VDWVOL - VOLUME INSIDE VAN-DER-WAALS SPHERE
REMARK SEVOL - VOLUME IN 1.4 ANGSTROM LAYER OUTSIDE VDW-SPHERE
REMARK BUR - INDICATES BURIED ATOMS (1=BURIED, 0=SURFACE)
REMARK
REMARK AT# ELEM RES C RES# X Y Z Z Z VDWVOL SEVOL BUR
"""
f_atom = 0
out = ""
mode = options['bfactor']
if mode == 'zscore':
z = ZScore(AveragePacking(options))
for l in open(infile).readlines():
if re.search('(^ATOM)|(^HETATM)',l):
if not f_atom:
f_atom = 1
if mode == 'packing_density': out += extralines_pd
elif mode == 'volume': out += extralines_vol
elif mode == 'zscore': out += extralines_z
# extract volume data
atom = self.parse_atom(l,options)
if check_valid and not atom['valid']: continue
if mode == 'packing_density': value = atom['packing_density']
elif mode == 'zscore':
value = z.get_z_score(atom,options)
if not value: value = -1
else: value = atom['total_volume']
newline = l[:55] + "%5.2f %5.2f"%(value,value) + l[67:]
out += newline
else:
out += l
open(outfile,'w').writelines(out+'\n')
def html_report(self,options):
report = """
<h1>Packing report</h1>
"""
total_atoms = len(self.atoms)
calc_atoms = 0
sum_pd = 0.0
for i in range(len(self.atoms)):
if self.valid[i]:
calc_atoms += 1
sum_pd += self.get_atom(i)['packing_density']
if calc_atoms > 0:
avg_pd = sum_pd / calc_atoms
else:
avg_pd = 0
n_cavi = len(self.cavities)
# calculate z-score-rms
avg_pack = AveragePacking(options)
z = ZScore(avg_pack)
z_score = z.compare_molecule(self,options)
report += """
<h2>1.SUMMARY</h2>
<table cellspacing=0 cellpadding=4 border=1>
<tr><td width='25%%'><b>structure file </b></td><td width='75%%'>%s</td></tr>
<tr><td><b>reference file </b></td><td>%s</td></tr>
<tr><td></td></tr>
<tr><td><b>total atoms in structure </b></td><td>%5i</td></tr>
<tr><td><b>atoms used for calculation </b></td><td>%5i</td></tr>
<tr><td><b>average packing density </b></td><td>%5.3f</td></tr>
<tr><td><b>number of cavities </b></td><td>%5i</td></tr>
<tr><td><b>z-score-rms </b></td><td>%5.2f</td></tr>
</table>
"""%(self.vol_file,options['reference_file'],total_atoms,calc_atoms,avg_pd,n_cavi,z_score)
report += """
<h2>2.ABBREVIATIONS USED</h2>
<table cellspacing=0 cellpadding=4 border=1>
<tr><td width='25%'><b> chain </b></td><td width='75%'> chain ID (one letter, e.g. H)</td></tr>
<tr><td><b> resi_no </b></td><td> residue number (integer, e.g. 131)</td></tr>
<tr><td><b> resi_type </b></td><td> residue type (e.g. TYR, HOH)</td></tr>
<tr><td><b> n_atoms_total </b></td><td> number of atoms in the structure</td></tr>
<tr><td><b> n_atoms </b></td><td> number of atoms used in the calculation</td></tr>
<tr><td><b> atom_no </b></td><td> atom number in PDB and .vol file.</td></tr>
<tr><td><b> atom_name </b></td><td> atom name (three-letter, from PDB file, e.g. CA, CD2)</td></tr>
<tr><td><b> atom_type </b></td><td> classes by which average packing is calculated.<br>
This is either 'residuename_atomname' e.g. 'TYR_CA', or one of 18 concise ProtOr atom types, e.g. 'C4H3u' for methyl groups.</td></tr>
<tr><td><b> <VdW> </b></td><td> average Van-der-Waals volume in cubic Angstroms</td></tr>
<tr><td><b> <SolvExc> </b></td><td> average Solvent Excluded volume in cubic Angstroms</td></tr>
<tr><td><b> <totVol> </b></td><td> average total volume (VdW + SolvExc) in cubic Angstroms</td></tr>
<tr><td><b> <PD> </b></td><td> average packing density (VdW/totVol)</td></tr>
<tr><td><b> SD </b></td><td> standard deviation</td></tr>
<tr><td><b> [Å<sup>3</sup>] </b></td><td> cubic Angstroms</td></tr>
</table>
<h2>3.ATOM NUMBERS AND PACKING</h2>
<table cellspacing=0 cellpadding=4 border=1>
<tr><th width='20%'>atomic subset</th><th width='5%'>n_atoms</th><th width='10%'><VdW> [Å<sup>3</sup>]</th><th width='10%%'><SolvExc> [Å<sup>3</sup>]</th><th width='10%'><totVol> [Å<sup>3</sup>]</th><th width='10%'>SD(<totVol>) [Å<sup>3</sup>]</th><th width='10%'><PD></th><th width='10%'>SD(<PD>)</th></tr>
"""
subsets = ["total atoms in structure ",
" peptidic atoms ",
" buried ",
" buried cavity neighbors ",
" other buried atoms ",
" surface ",
" surface cavity neighbors ",
" other surface atoms ",
" hetero atoms ",
" water molecules ",
" other hetero atoms "]
avg_pack = AveragePacking(options)
for s in subsets:
rec = avg_pack.count_subset(self,re.sub('(^\s+)|(\s+\Z)','',s))
line = '<tr align="right"><td align="left">%s</td><td>%i</td><td>%5.2f</td><td>%5.2f</td><td>%5.2f</td><td>%5.2f</td><td>%5.3f</td><td>%5.3f</td></tr>\n'%(s,rec[0],rec[1],rec[2],rec[3],rec[4],rec[5],rec[6])
report += line
report += """</table>
<h2>4.CAVITIES</h2>
<table cellspacing=0 cellpadding=4 border=1>
<tr><th width='15%'>cavity_no</th><th width='10%'>n_neighbors</th><th width='10%'><dist> [Å]</th><th width='10%'>%surface</th><th width='30%'>atom_types</th><th width='20%'>atom_indices</th></tr>
"""
for i in range(len(self.cavities)):
cav = self.cavities[i]
nsurf = 0
for c in cav:
if self.get_atom(c)['surface']: nsurf +=1
atypes = string.join([avg_pack.get_atomtype(self.get_atom(ai)) for ai in cav],', ')
anums = string.join([str(num) for num in cav],', ')
eu_med = self.cav_dist[i]
report += '<tr align="right"><td>%3i</td><td>%3i</td><td>%5.2f</td><td>%4.1f</td><td align="left">%s</td><td align="left">%s</td></tr>\n'%(i,len(cav),eu_med,nsurf*100.0/len(cav),atypes,anums)
report += """</table>
<h2>5.PACKING PER RESIDUE</h2>
<table cellspacing=0 cellpadding=4 border=1>
<tr><th width='5%'>chain</th><th width='10%'>resi_no</th><th width='10%'>resi_type</th><th width='5%'>n_atoms</th><th width='10%'>sum(VdW) [Å<sup>3</sup>]</th><th width='10%'>sum(SolvExc) [Å<sup>3</sup>]</th><th width='10%'>sum(totVol) [Å<sup>3</sup>]</th><th width='10%'><PD></th><th width='10%'>SD(<PD>)</th></tr>
"""
avg_pack = AveragePacking(options)
for i in range(len(self.atoms)):
if self.valid[i]:
atom = self.get_atom(i)
avg_pack.count_residue(atom)
for r in avg_pack.get_residue_results():
report += '<tr align="right"><td>%s</td><td>%s</td><td>%s</td><td>%i</td><td>%5.2f</td><td>%5.2f</td><td>%5.2f</td><td>%5.2f</td><td>%5.3f</td></tr>\n'%r
report += """</table>
<h2>6.PACKING PER ATOM</h2>
<table cellspacing=0 cellpadding=4 border=1>
<tr><th width='10%'>atom_no</th><th width='5%'>chain</th><th width='10%'>resi_no</th><th width='10%'>residue_name</th><th width='10%'>atom_name</th><th width='10%'>atom_type</th><th width='10%'>VdWVol [Å<sup>3</sup>]</th><th width='10%'>SolvExcVol [Å<sup>3</sup>]</th><th width='10%'>totVol [Å<sup>3</sup>]</th><th width='10%'>PD</th></tr>
"""
for i in range(len(self.atoms)):
if self.valid[i]:
atom = self.get_atom(i)
atomtype = avg_pack.get_atomtype(atom)
report += '<tr align="right"><td>%i</td><td>%s</td><td>%i</td><td>%s</td><td>%s</td><td>%s</td><td>%5.2f</td><td>%5.2f</td><td>%5.2f</td><td>%5.3f</td></tr>\n'%(atom['index'],atom['chain_id'],atom['residue_number'],atom['residue_type'],atom['atom_type'],atomtype,atom['vdw_volume'],atom['solv_ex_volume'],atom['total_volume'],atom['packing_density'])
report += """</table>"""
return report
