def curate_homoDB(verbosity):
'''
Creates h**o-oligomeric database from a local pdb repsitory.
The divided scheme adopted by RCSB, in which the subdirectories
are the two middle characters in the PDB code, is assumed.
Each database contains three key files: dat, log and fasta.
* homodb.dat contains only the pdb codes contained in the database.
* homodb.log contains summarized relevant information about each entry.
* homodb.fasta contains the sequences of every chain in the database.
Called by: update_databases()
'''
# Create stats folder if does not exist
stats_dir = os.path.join(pdb_homo_archive, 'stats')
if not os.path.isdir(stats_dir):
os.mkdir(stats_dir)
# Compare latest assession with new files
assession_log = read_latest_assession(stats_dir)
new_files = list_new_files(pdb1_archive, assession_log, verbosity)
print(clrs['g'] + str(len(new_files)) + clrs['n'] +
' new structure files were found and will be processed')
now = str(time.strftime("%d-%m-%Y@%H.%M.%S"))
dat_file = os.path.join(stats_dir, now + '-choirdb.dat')
log_file = os.path.join(stats_dir, now + '-choirdb.log')
err_file = os.path.join(stats_dir, now + '-choirdb.err')
if not os.path.isfile(dat_file):
with open(dat_file, 'w+'):
pass
# Write files not to be updated to new dat file
with open(dat_file, 'a') as f:
for i in assession_log:
if i not in new_files:
f.write(i + " " + assession_log[i] + "\n")
# Create log file
if not os.path.isfile(log_file):
with open(log_file, 'w+') as f:
f.write('Code, Chains, Author, Software, Date\n')
# Read Chain correspondences
chain_correspondences_file = os.path.join(stats_dir,
'chain_correspondences.pickle')
if os.path.isfile(chain_correspondences_file):
with open(chain_correspondences_file, 'rb') as p:
chain_correspondences = pickle.load(p)
else:
chain_correspondences = {}
# Main loop that will populate the ProtCHOIR database
for pdb in pg(new_files, widgets=widgets):
filename = pdb.split('/')[-1]
subfolder = pdb.split('/')[-2]
# Record assessment in dat file
with open(dat_file, 'a') as f:
f.write(filename + " " + str(time.time()) + '\n')
# Start assession
pctools.printv('\nAssessing ' + pdb + '...', verbosity)
# Reject files larger than 10Mb
file_size = os.stat(pdb).st_size / 1048576
pctools.printv(
'File size: ' + clrs['c'] + '{0:.1g}'.format(file_size) + ' Mb' +
clrs['n'], verbosity)
if file_size > 2:
pctools.printv(clrs['r'] + "File size too large!" + clrs['n'],
verbosity)
pctools.printv(
clrs['y'] +
"Will try to fetch sequences from asymmetric unit." +
clrs['n'], verbosity)
try:
alternative_pdb = os.path.join(
pdb_archive, subfolder,
'pdb' + filename.split('.')[0] + '.ent.gz')
pdb_code, structure, nchains = pctools.parse_pdb_structure(
alternative_pdb)
structure, chain_correspondences[
pdb_code] = pctools.split_states(structure)
nchainspostsplit, seqs, chain_ids = pctools.extract_seqs(
structure, 0)
# Write in fasta file
pctools.printv(
clrs['y'] + "Recording large-pdb sequence" + clrs['n'],
verbosity)
record_fasta(pdb_code,
seqs,
chain_ids,
subfolder,
type='largepdb')
except:
pctools.printv(
clrs['r'] + "Failed to fetch sequence!" + clrs['n'],
verbosity)
continue
try:
pdb_code, structure, nchains = pctools.parse_pdb_structure(pdb)
pctools.printv(
'Number of chains in structure ' + clrs['y'] + pdb_code +
clrs['n'] + ': ' + str(nchains), verbosity)
# Reject structures with more than 60 chains
if int(nchains) > 60:
pctools.printv(
"Number of chains (" + clrs['y'] + str(nchains) +
clrs['n'] + ") larger than 60! " + clrs['r'] +
"Too many chains!" + clrs['n'], verbosity)
pctools.printv(
clrs['y'] + "Will try to fetch sequences anyway." +
clrs['n'], verbosity)
try:
pdb_code, structure, nchains = pctools.parse_pdb_structure(
pdb)
structure, chain_correspondences[
pdb_code] = pctools.split_states(structure)
nchainspostsplit, seqs, chain_ids = pctools.extract_seqs(
structure, 0)
pctools.printv(
clrs['y'] + "Recording large-pdb sequence" + clrs['n'],
verbosity)
# Write in fasta file
record_fasta(pdb_code,
seqs,
chain_ids,
subfolder,
type='largepdb')
except:
pctools.printv(
clrs['r'] + "Failed to fetch sequence!" + clrs['n'],
verbosity)
continue
structure, chain_correspondences[pdb_code] = pctools.split_states(
structure)
nchainspostsplit, seqs, chain_ids = pctools.extract_seqs(
structure, 0)
pctools.printv(
'Number of chains (' + clrs['c'] + str(nchains) + clrs['n'] +
') and file size (' + clrs['c'] + str(file_size) + clrs['n'] +
') OK.' + clrs['g'] + ' Proceeding.' + clrs['n'] + '\n',
verbosity)
# Try to get info from the canonic pdb header (homonimous to pdb1)
canonpdb = "pdb" + pdb_code + ".ent.gz"
try:
contents = pctools.parse_pdb_contents(
os.path.join(pdb_archive, subfolder, canonpdb))[1]
except:
pctools.printv(
clrs['r'] +
'\n\n Mismatch between pdb and biounit entries...' +
clrs['n'], verbosity)
author, software = pctools.get_annotated_states(contents)
pctools.printv(
'Author determined biological unit = ' + str(author),
verbosity)
pctools.printv(
'Software determined quaternary structure= ' + str(software),
verbosity)
# Start assessing sequences and structures (from 2 up to 26 chains)
if 1 < int(nchains) < 61:
ids, proteinpair = pctools.get_pairwise_ids(seqs, nchains)
for id in ids:
if id[0] >= 90:
color = clrs['g']
else:
color = clrs['r']
pctools.printv(
'Identity between chains ' + clrs['y'] + str(id[1]) +
clrs['n'] + ' and ' + clrs['y'] + str(id[2]) +
clrs['n'] + ' is ' + color + str(id[0]) + "%" +
clrs['n'] + ".", verbosity)
# Save records for pure h**o-oligomers
if all(id[0] > 90 for id in ids) and proteinpair is True:
pctools.printv(
"All identities over 90%. Likely " + clrs['b'] +
"h**o-oligomeric" + clrs['n'] + ".", verbosity)
pctools.printv(clrs['y'] + "FETCHING" + clrs['n'] + ".\n",
verbosity)
# Write file to database
newfile = os.path.join(pdb_homo_archive, subfolder,
pdb_code + ".pdb")
if not os.path.isdir(
os.path.join(pdb_homo_archive, subfolder)):
os.mkdir(os.path.join(pdb_homo_archive, subfolder))
io.set_structure(structure)
io.save(newfile)
pctools.gzip_pdb(newfile)
# Write to log file
with open(log_file, 'a') as f:
f.write(
str(pdb_code) + "," + str(nchains) + "," +
'/'.join(author) + "," + '/'.join(software) + "," +
str(os.path.getctime(newfile + '.gz')) + '\n')
# Write in fasta file
pctools.printv(
clrs['y'] + "Recording h**o-oligomer sequence." +
clrs['n'], verbosity)
record_fasta(pdb_code,
seqs,
chain_ids,
subfolder,
type='h**o')
# Investigate partial h**o-oligomers
elif any(id[0] > 90 for id in ids) and proteinpair is True:
at_least_one_interface = False
for id in ids:
if id[0] > 90:
# Check if similar chains share interfaces
if pctools.check_interfaces(
structure, id[1], id[2]):
at_least_one_interface = True
pctools.printv(
'Contacts found between chains ' +
clrs['g'] + str(id[1]) + clrs['n'] +
' and ' + clrs['g'] + str(id[2]) +
clrs['n'] + ' sharing ' + clrs['g'] +
str(id[0]) + clrs['n'] + " % identity.",
verbosity)
pctools.printv(
"At least one putative " + clrs['b'] +
"h**o-oligomeric " + clrs['n'] +
"interface found.", verbosity)
pctools.printv(
clrs['y'] + "FETCHING" + clrs['n'] + ".\n",
verbosity)
# Write file to database
newfile = os.path.join(pdb_homo_archive,
subfolder,
pdb_code + ".pdb")
if not os.path.isdir(
os.path.join(pdb_homo_archive,
subfolder)):
os.mkdir(
os.path.join(pdb_homo_archive,
subfolder))
io.set_structure(structure)
io.save(newfile)
pctools.gzip_pdb(newfile)
# Write to log file
with open(log_file, 'a') as f:
f.write(
str(pdb_code) + "," + str(nchains) +
"," + '/'.join(author) + "," +
'/'.join(software) + "," +
str(os.path.getctime(newfile +
'.gz')) + '\n')
# Write in fasta file
pctools.printv(
clrs['y'] +
"Recording h**o-oligomer sequence." +
clrs['n'], verbosity)
record_fasta(pdb_code,
seqs,
chain_ids,
subfolder,
type='h**o')
break
if at_least_one_interface is False:
pctools.printv(
"No h**o-oligomeric interface found. Likely " +
clrs['r'] + "hetero-oligomeric" + clrs['n'] + ".",
verbosity)
pctools.printv(
clrs['y'] + "Recording hetero-oligomer sequence" +
clrs['n'], verbosity)
# Write in fasta file
record_fasta(pdb_code,
seqs,
chain_ids,
subfolder,
type='hetero')
elif proteinpair is False:
pctools.printv(
clrs['r'] + "No proteic chain pairs found" +
clrs['n'] + ".", verbosity)
if any([set(seq[1]) != {'X'} for seq in seqs]):
pctools.printv(
clrs['y'] + "Protein sequences found though" +
clrs['n'], verbosity)
pctools.printv(
clrs['y'] + "Recording hetero-oligomer sequence" +
clrs['n'], verbosity)
# Write in fasta file
record_fasta(pdb_code,
seqs,
chain_ids,
subfolder,
type='hetero')
else:
pctools.printv(
clrs['r'] +
"Not even a single protein chain. Disregarding." +
clrs['n'], verbosity)
else:
pctools.printv(
"No similar chains found. Likely " + clrs['r'] +
"hetero-oligomeric" + clrs['n'] + ".", verbosity)
pctools.printv(
clrs['y'] + "Recording hetero-oligomer sequence" +
clrs['n'], verbosity)
record_fasta(pdb_code,
seqs,
chain_ids,
subfolder,
type='hetero')
elif int(nchains) == 1:
pctools.printv(
"Only one chain found. Likely " + clrs['r'] + "monomeric" +
clrs['n'] + ".", verbosity)
pctools.printv(
clrs['y'] + "Recording monomer sequence." + clrs['n'],
verbosity)
structure, chain_correspondences[
pdb_code] = pctools.split_states(structure)
nchains, seqs, chain_ids = pctools.extract_seqs(structure, 0)
record_fasta(pdb_code, seqs, chain_ids, subfolder, type='mono')
except:
errtype, errvalue, errtraceback = sys.exc_info()
errtypeshort = str(errtype).split('\'')[1]
pctools.printv(
clrs['r'] + '*' + str(errtypeshort) + ': ' + str(errvalue) +
' l.' + str(errtraceback.tb_lineno) + '*' + clrs['n'],
verbosity)
traceback.print_exception(*sys.exc_info())
if errtypeshort == 'KeyboardInterrupt':
quit()
#pctools.printv(clrs['r']+"UNKNOWN FAULT"+clrs['n']+".", verbosity)
if not os.path.isfile(err_file):
with open(err_file, 'w+') as f:
pass
with open(err_file, 'a') as f:
f.write(filename + '\n')
continue
with open(chain_correspondences_file, 'wb') as p:
pickle.dump(chain_correspondences, p, protocol=pickle.HIGHEST_PROTOCOL)
if not os.path.isfile(err_file):
with open(err_file, 'w+') as f:
f.write('\nNo errors. Assessment terminated succesfully.\n')
def analyse_model(oligomer):
output = []
model_report = g_report.copy()
model_report['model_filename'] = oligomer
model_oligomer_name = os.path.basename(oligomer).split(
"_CHOIR_")[0].replace('.', '_')
output.append(pctools.subsection('3', model_oligomer_name))
output.append('Analysing oligomer file: ' + clrs['y'] + oligomer +
clrs['n'] + '\n')
model_report['model_oligomer_name'] = model_oligomer_name
if g_args.generate_report is True:
model_report['model_figures'], pymol_output = pctools.pymol_screenshot(
oligomer, g_args, putty=True)
output.append(pymol_output)
pdb_name, structure, nchains = pctools.parse_any_structure(oligomer)
nchains, seqs, chain_ids = pctools.extract_seqs(structure, 0)
relevant_chains = []
for seq in seqs:
relevant_chains.append(seq[0])
pisa_output, pisa_error, protomer_data = pctools.run_pisa(
oligomer,
'',
g_args.verbosity,
gen_monomer_data=True,
gen_oligomer_data=True)
protomer_surface_residues = pctools.get_areas(protomer_data)
model_report['assemblied_protomer_plot'], model_report[
'assemblied_protomer_exposed_area'], model_report[
'assemblied_protomer_hydrophobic_area'], model_report[
'assemblied_protomer_conserved_area'], minx, maxx, analysis_output = pctools.plot_analysis(
pdb_name,
protomer_surface_residues,
g_entropies,
g_z_entropies,
g_tmdata,
g_args,
minx=g_minx,
maxx=g_maxx)
output.append(analysis_output)
if 'I' in g_args.assessment and not g_args.allow_monomers:
output.append(
pctools.subsection('3' + '[I]', 'Interfaces Comparison: ' +
model_oligomer_name))
if g_args.sequence_mode is False and g_args.skip_conservation is False:
model_report['exposed_area_reduction'] = int(
100 *
(float(model_report['assemblied_protomer_exposed_area']) -
float(model_report['protomer_exposed_area'])) /
float(model_report['protomer_exposed_area']))
model_report['hydrophobic_area_reduction'] = int(
100 *
(float(model_report['assemblied_protomer_hydrophobic_area']) -
float(model_report['protomer_hydrophobic_area'])) /
float(model_report['protomer_hydrophobic_area']))
model_report['conserved_area_reduction'] = int(
100 *
(float(model_report['assemblied_protomer_conserved_area']) -
float(model_report['protomer_conserved_area'])) /
float(model_report['protomer_conserved_area']))
if model_report['exposed_area_reduction'] < -5:
if model_report['hydrophobic_area_reduction'] < 0:
hydophobic_surface_score = 10 * (
model_report['hydrophobic_area_reduction'] /
model_report['exposed_area_reduction']) / 3
else:
hydophobic_surface_score = 0
if hydophobic_surface_score > 10:
hydophobic_surface_score = 10
output.append('Hydrophobic surface score: ' +
str(hydophobic_surface_score))
if model_report['conserved_area_reduction'] < 0:
conserved_surface_score = 10 * (
model_report['conserved_area_reduction'] /
model_report['exposed_area_reduction']) / 3
else:
conserved_surface_score = 0
if conserved_surface_score > 10:
conserved_surface_score = 10
output.append('Conserved surface score: ' +
str(conserved_surface_score))
model_report['surface_score'] = round(
(hydophobic_surface_score + conserved_surface_score) / 2,
2)
else:
output.append(clrs['r'] + 'Exposed area reduction too small.' +
clrs['n'])
model_report['surface_score'] = 0
output.append('Final surface score: ' +
str(model_report['surface_score']))
else:
model_report['surface_score'] = 'NA'
model_oligomer = oligomer.split('_CHOIR_CorrectedChains')[0]
xml_out = model_oligomer + '_CHOIR_PisaInterfaces.xml'
model_interfaces_list, interfaces_output = pctools.parse_interfaces(
xml_out, relevant_chains, g_args.verbosity)
template_interfaces_list = g_interfaces_dict[g_template_hitchain]
if model_interfaces_list and template_interfaces_list:
if g_args.verbosity > 0:
output.append(clrs['y'] + 'MODEL INTERFACES' + clrs['n'])
for model_interface in model_interfaces_list:
output.append(clrs['y'] +
' <> '.join(model_interface['chains']) +
clrs['n'])
output.append(clrs['y'] + 'Interface Area: ' + clrs['n'] +
str(model_interface['interface area']) +
' A^2')
output.append(
clrs['y'] + 'Interface Solvation Energy: ' +
clrs['n'] +
str(model_interface['interface solvation energy']) +
' kcal/mol')
output.append(clrs['y'] + 'Hydrogen Bonds: ' + clrs['n'] +
str(model_interface['hydrogen bonds']))
output.append(clrs['y'] + 'Salt Bridges: ' + clrs['n'] +
str(model_interface['salt bridges']))
output.append(clrs['y'] + 'Disulphide Bridges: ' +
clrs['n'] +
str(model_interface['disulphide bridges']) +
"\n\n")
interfaces_comparison = {}
for template_interface in template_interfaces_list:
for model_interface in model_interfaces_list:
if set(model_interface['chains']) == set(
template_interface['chains']):
comparison_data = {}
denominator = 12
delta_area = round(
model_interface['interface area'] -
template_interface['interface area'], 2)
comparison_data['model area'] = model_interface[
'interface area']
comparison_data['template area'] = template_interface[
'interface area']
comparison_data['delta area'] = delta_area
delta_energy = round(
model_interface['interface solvation energy'] -
template_interface['interface solvation energy'],
2)
comparison_data['model energy'] = model_interface[
'interface solvation energy']
comparison_data[
'template energy'] = template_interface[
'interface solvation energy']
comparison_data['delta energy'] = delta_energy
delta_hb = round(
model_interface['hydrogen bonds'] -
template_interface['hydrogen bonds'], 2)
comparison_data['model hb'] = model_interface[
'hydrogen bonds']
comparison_data['template hb'] = template_interface[
'hydrogen bonds']
comparison_data['delta hb'] = delta_hb
delta_sb = round(
model_interface['salt bridges'] -
template_interface['salt bridges'], 2)
comparison_data['model sb'] = model_interface[
'salt bridges']
comparison_data['template sb'] = template_interface[
'salt bridges']
comparison_data['delta sb'] = delta_sb
delta_ss = round(
model_interface['disulphide bridges'] -
template_interface['disulphide bridges'], 2)
comparison_data['model ss'] = model_interface[
'disulphide bridges']
comparison_data['template ss'] = template_interface[
'disulphide bridges']
comparison_data['delta ss'] = delta_ss
output.append(clrs['y'] + 'INTERFACES COMPARISON' +
clrs['n'])
output.append(' <> '.join(model_interface['chains']))
if delta_area >= 0:
emphasis_color = clrs['g']
relative_area = 100
else:
emphasis_color = clrs['r']
relative_area = round(
model_interface['interface area'] * 100 /
template_interface['interface area'], 2)
output.append('Delta Interface Area: ' +
emphasis_color + str(delta_area) +
clrs['n'] + ' A^2 (' +
str(relative_area) + '%)')
if delta_energy <= 0:
emphasis_color = clrs['g']
relative_energy = 100
else:
emphasis_color = clrs['r']
if model_interface[
'interface solvation energy'] < 0 and template_interface[
'interface solvation energy'] < 0:
relative_energy = round(
model_interface[
'interface solvation energy'] * 100 /
template_interface[
'interface solvation energy'], 2)
elif model_interface[
'interface solvation energy'] > 0 and template_interface[
'interface solvation energy'] < 0:
relative_energy = 0
elif model_interface[
'interface solvation energy'] < 0 and template_interface[
'interface solvation energy'] > 0:
relative_energy = 100
elif model_interface[
'interface solvation energy'] > 0 and template_interface[
'interface solvation energy'] > 0:
relative_energy = 0
output.append('Delta Interface Solvation Energy: ' +
emphasis_color + str(delta_energy) +
clrs['n'] + ' kcal/mol (' +
str(relative_energy) + '%)')
if model_interface[
'hydrogen bonds'] == template_interface[
'hydrogen bonds'] == 0:
relative_hb = 0
emphasis_color = clrs['r']
denominator -= 2
elif delta_hb >= 0:
relative_hb = 100
emphasis_color = clrs['g']
else:
emphasis_color = clrs['r']
relative_hb = round(
model_interface['hydrogen bonds'] * 100 /
template_interface['hydrogen bonds'], 2)
output.append('Delta Hydrogen Bonds: ' +
emphasis_color + str(delta_hb) +
clrs['n'] + ' (' + str(relative_hb) +
'%)')
if model_interface[
'salt bridges'] == template_interface[
'salt bridges'] == 0:
relative_sb = 0
emphasis_color = clrs['r']
denominator -= 3
elif delta_sb >= 0:
relative_sb = 100
emphasis_color = clrs['g']
else:
relative_sb = round(
model_interface['salt bridges'] * 100 /
template_interface['salt bridges'], 2)
emphasis_color = clrs['r']
output.append('Delta Salt Bridges: ' + emphasis_color +
str(delta_sb) + clrs['n'] + ' (' +
str(relative_sb) + '%)')
if model_interface[
'disulphide bridges'] == template_interface[
'disulphide bridges'] == 0:
relative_ss = 0
emphasis_color = clrs['r']
denominator -= 4
elif delta_ss >= 0:
relative_ss = 100
emphasis_color = clrs['g']
else:
relative_ss = round(
model_interface['disulphide bridges'] * 100 /
template_interface['disulphide bridges'], 2)
emphasis_color = clrs['r']
output.append('Delta Disulphide Bridges: ' +
emphasis_color + str(delta_ss) +
clrs['n'] + ' (' + str(relative_ss) +
'%)\n')
if denominator == 0:
comparison_data['score'] = 0
else:
comparison_data['score'] = round(
(relative_area + 2 * relative_energy +
2 * relative_hb + 3 * relative_sb +
4 * relative_ss) / denominator, 2)
output.append('Interface score: ' +
str(comparison_data['score']))
interfaces_comparison[''.join(
sorted(
model_interface['chains']))] = comparison_data
comparison_plots, interfaces_output = plot_deltas(
model_oligomer_name, template, interfaces_comparison, g_args)
model_report['comparison_plots'] = os.path.basename(
comparison_plots)
output.append(interfaces_output)
summed_score = 0
for interface, data in interfaces_comparison.items():
summed_score += data['score']
model_report['interfaces_score'] = round(
summed_score / (10 * len(interfaces_comparison)), 2)
output.append('Final interfaces score: ' +
str(model_report['interfaces_score']))
else:
if 'surface_score' not in model_report:
model_report['surface_score'] = 0
model_report['interfaces_score'] = 0
else:
model_report['surface_score'] = 'NA'
model_report['interfaces_score'] = 'NA'
model_report['comparison_plots'] = 'NA'
model_report['assemblied_protomer_exposed_area'] = 'NA'
model_report['assemblied_protomer_hydrophobic_area'] = 'NA'
model_report['assemblied_protomer_conserved_area'] = 'NA'
if 'G' in g_args.assessment:
output.append(pctools.subsection('3' + '[G]', 'GESAMT Comparison'))
qscore, rmsd, fasta_out, gesamt_output = pctools.run_gesamt(
template, template_file, model_oligomer_name, oligomer, None,
g_args)
output.append(gesamt_output)
model_report['gesamt_qscore'] = str(qscore)
model_report['gesamt_rmsd'] = str(rmsd)
else:
model_report['gesamt_qscore'] = 'NA'
model_report['gesamt_rmsd'] = 'NA'
if 'M' in g_args.assessment:
output.append(pctools.subsection('3' + '[M]', 'Molprobity Comparison'))
model_molprobity, molprobity_output = pctools.run_molprobity(
oligomer, g_args)
output.append(molprobity_output)
model_report['model_clashscore'] = str(model_molprobity['clashscore'])
model_report['model_molprobity'] = str(
model_molprobity['molprobity_score'])
output.append(clrs['y'] + 'MOLPROBITY COMPARISON' + clrs['n'])
output.append('Criterion\tTempl.\tModel')
output.append('Rama. Fav.\t' + str(template_molprobity['rama_fav']) +
'\t' + str(model_molprobity['rama_fav']))
output.append('Rama. Out.\t' + str(template_molprobity['rama_out']) +
'\t' + str(model_molprobity['rama_out']))
output.append('Rot. Out.\t' + str(template_molprobity['rot_out']) +
'\t' + str(model_molprobity['rot_out']))
output.append('CBeta Dev.\t' + str(template_molprobity['cb_dev']) +
'\t' + str(model_molprobity['cb_dev']))
output.append('Clashscore\t' + str(template_molprobity['clashscore']) +
'\t' + str(model_molprobity['clashscore']))
output.append('Molprob. Score\t' +
str(template_molprobity['molprobity_score']) + '\t' +
str(model_molprobity['molprobity_score']))
molprobity_radar, radar_output = plot_molprobity(
model_oligomer_name, model_molprobity, template,
template_molprobity)
output.append(radar_output)
model_report['molprobity_radar'] = molprobity_radar
delta_clashscore = (model_molprobity['clashscore'] -
template_molprobity['clashscore']) / 10
output.append('Delta clashscore: ' + str(delta_clashscore))
if delta_clashscore >= 1:
model_report['quality_score'] = round(
10 - math.log(delta_clashscore**5, 10), 2)
else:
model_report['quality_score'] = 10
output.append('Final quality score: ' +
str(model_report['quality_score']))
else:
model_report['model_clashscore'] = 'NA'
model_report['model_molprobity'] = 'NA'
model_report['quality_score'] = 'NA'
if 'M' in g_args.assessment and 'I' in g_args.assessment and not g_args.allow_monomers:
if g_args.sequence_mode is False and g_args.skip_conservation is False:
model_report['protchoir_score'] = round(
sum([
model_report['interfaces_score'],
model_report['surface_score'],
model_report['quality_score']
]) / 3, 2)
else:
model_report['protchoir_score'] = round(
sum([
model_report['interfaces_score'],
model_report['quality_score']
]) / 2, 2)
elif 'M' in g_args.assessment:
model_report['protchoir_score'] = model_report['quality_score']
elif 'I' in g_args.assessment:
if g_args.sequence_mode is False and g_args.skip_conservation is False:
model_report['protchoir_score'] = round(
sum([
model_report['interfaces_score'],
model_report['surface_score']
]) / 2, 2)
else:
model_report['protchoir_score'] = model_report['interfaces_score']
else:
model_report['protchoir_score'] = 'NA'
if str(model_report['protchoir_score']) == 'NA':
model_report['score_color'] = 'grey'
elif model_report['protchoir_score'] <= 5:
model_report['score_color'] = 'red'
elif 5 < model_report['protchoir_score'] <= 7:
model_report['score_color'] = 'orange'
elif model_report['protchoir_score'] > 7:
model_report['score_color'] = 'green'
pickle.dump(model_report,
open(model_oligomer_name + '_CHOIR_model_report.pickle', 'wb'))
return model_report, '\n'.join(output)