Skip to content

Commit

Permalink
make only multiple models usable #46
Browse files Browse the repository at this point in the history
  • Loading branch information
@famosab
famosab committed Mar 7, 2023
1 parent ef9b9bd commit 9b21b73
Show file tree
Hide file tree
Showing 4 changed files with 205 additions and 194 deletions.
8 changes: 4 additions & 4 deletions .github/workflows/docs.yml
View file
Original file line number Diff line number Diff line change
Expand Up @@ -2,10 +2,10 @@ name: Update Docs

on:
workflow_dispatch:
# push:
# paths:
# - '**.rst'
# - 'docs/*'
push:
paths:
- '**.rst'
- 'docs/*'

jobs:
run:
Expand Down
1 change: 1 addition & 0 deletions config.yaml
View file
Original file line number Diff line number Diff line change
Expand Up @@ -25,6 +25,7 @@ multiple_paths: # enter as many paths as you need below
- 'data/e_coli_core.xml'
- ''
- ''
single: TRUE # set to False if you only want to work with the multiple models

# media to simulate growth from, just comment the media you do not want with a #
media:
Expand Down
234 changes: 119 additions & 115 deletions main.py
View file
Original file line number Diff line number Diff line change
Expand Up @@ -34,6 +34,9 @@ def main(configpath=None):
logging.StreamHandler()
]
)
logging.getLogger('cobra').setLevel(logging.WARNING)
logging.getLogger('requests').setLevel(logging.WARNING)
logging.getLogger('urllib3').setLevel(logging.WARNING)
logging.info('----------- New run of refineGEMs -----------')
rg.databases.initialise_database()

Expand Down Expand Up @@ -70,134 +73,135 @@ def main(configpath=None):
heatmap.savefig(config['out_path'] + 'visualization/' + 'heatmap_dt_additives_' + str(today) + '.png')
binary_heatmap.savefig(config['out_path'] + 'visualization/' + 'heatmap_native_' + str(today) + '.png', bbox_inches='tight')
ini_plot.savefig(config['out_path'] + 'visualization/' + 'model_status_' + str(today) + '.png', bbox_inches='tight')

try:
model_cobra, errors = cobra.io.sbml.validate_sbml_model(config['model'])
logging.info(errors)
except (OSError):
model_cobra = None
logging.info('No or no valid model given, please enter a valid path in the model field in the config file.')

if (config['single']):
try:
model_cobra, errors = cobra.io.sbml.validate_sbml_model(config['model'])
logging.info(errors)
except (OSError):
model_cobra = None
logging.info('No or no valid model given, please enter a valid path in the model field in the config file.')

if (config['keggpathways']):
model_libsbml, non_kegg = rg.pathways.kegg_pathways(config['model'])
file = open(config['out_path'] + model_libsbml.getId() + '_reac_wo_kegg_' + str(today) + '.txt','w')
for reaction in non_kegg:
file.write(reaction+"\n")
file.close()
logging.info('Kegg Pathways were added to the model as groups. Reactions that have no KEGG annotation are denoted in ' + model_libsbml.getId() + '_reac_wo_kegg.txt')
if (config['keggpathways']):
model_libsbml, non_kegg = rg.pathways.kegg_pathways(config['model'])
file = open(config['out_path'] + model_libsbml.getId() + '_reac_wo_kegg_' + str(today) + '.txt','w')
for reaction in non_kegg:
file.write(reaction+"\n")
file.close()
logging.info('Kegg Pathways were added to the model as groups. Reactions that have no KEGG annotation are denoted in ' + model_libsbml.getId() + '_reac_wo_kegg.txt')

else:
model_libsbml = rg.io.load_model_libsbml(config['model'])

if (config['sboterms']):
if (config['visualize']):
sbo_fig = rg.investigate.plot_rea_sbo_single(model_libsbml).get_figure()
# saving the created visualizations
sbo_fig.savefig(config['out_path'] + 'visualization/' + str(model_cobra.id) + '_ReacPerSBO_beforeUpdate_' + str(today) + '.png', bbox_inches='tight')
model_libsbml = rg.sboann.sbo_annotation(model_libsbml)
logging.info('SBO Terms updated for ' + model_libsbml.getId())

if (config['charge_corr']):
model_libsbml, multiple_charges = rg.charges.correct_charges_modelseed(model_libsbml)
pd.DataFrame.from_dict(multiple_charges, orient='index').to_csv(config['out_path'] + model_libsbml.getId() + '_mulchar_' + str(today) + '.csv', sep=',', header=False)
logging.info('Charges were corrected for ' + model_libsbml.getId() + '. A table with metabolites with multiple charges can be found under ' + model_libsbml.getId() + '_mulchar_' + str(today) + '.csv')
else:
model_libsbml = rg.io.load_model_libsbml(config['model'])

if(config['man_cur']):
if config['man_cur_type'] == 'gapfill':
gapfill = rg.io.load_manual_gapfill(config['man_cur_table'])
model_libsbml = rg.curate.add_reactions_from_table(model_libsbml, gapfill, config['entrez_email'])
logging.info('Manual gap filling was done for ' + model_libsbml.getId())
elif config['man_cur_type'] == 'metabs':
man_ann = rg.io.load_manual_annotations(config['man_cur_table'])
model_libsbml = rg.curate.update_annotations_from_table(model_libsbml, man_ann)
model_libsbml = rg.curate.update_annotations_from_others(model_libsbml)
logging.info('Manual update of annotations was done for ' + model_libsbml.getId())
if (config['sboterms']):
if (config['visualize']):
sbo_fig = rg.investigate.plot_rea_sbo_single(model_libsbml).get_figure()
# saving the created visualizations
sbo_fig.savefig(config['out_path'] + 'visualization/' + str(model_cobra.id) + '_ReacPerSBO_beforeUpdate_' + str(today) + '.png', bbox_inches='tight')
model_libsbml = rg.sboann.sbo_annotation(model_libsbml)
logging.info('SBO Terms updated for ' + model_libsbml.getId())

if config['gap_analysis'] and config['gapfill_model']:
filename = f'{config["out_path"]}{name}_gap_analysis_{str(today)}.xlsx'
if config['gap_analysis_params'].get('db_to_compare') not in ('BioCyc', 'KEGG+BioCyc'):
logging.warning('Currently, only the result from the \'BioCyc\' or \'KEGG+BioCyc\' runs can be directly added to a model.')
if (config['charge_corr']):
model_libsbml, multiple_charges = rg.charges.correct_charges_modelseed(model_libsbml)
pd.DataFrame.from_dict(multiple_charges, orient='index').to_csv(config['out_path'] + model_libsbml.getId() + '_mulchar_' + str(today) + '.csv', sep=',', header=False)
logging.info('Charges were corrected for ' + model_libsbml.getId() + '. A table with metabolites with multiple charges can be found under ' + model_libsbml.getId() + '_mulchar_' + str(today) + '.csv')

if(config['man_cur']):
if config['man_cur_type'] == 'gapfill':
gapfill = rg.io.load_manual_gapfill(config['man_cur_table'])
model_libsbml = rg.curate.add_reactions_from_table(model_libsbml, gapfill, config['entrez_email'])
logging.info('Manual gap filling was done for ' + model_libsbml.getId())
elif config['man_cur_type'] == 'metabs':
man_ann = rg.io.load_manual_annotations(config['man_cur_table'])
model_libsbml = rg.curate.update_annotations_from_table(model_libsbml, man_ann)
model_libsbml = rg.curate.update_annotations_from_others(model_libsbml)
logging.info('Manual update of annotations was done for ' + model_libsbml.getId())

if config['gap_analysis'] and config['gapfill_model']:
filename = f'{config["out_path"]}{name}_gap_analysis_{str(today)}.xlsx'
if config['gap_analysis_params'].get('db_to_compare') not in ('BioCyc', 'KEGG+BioCyc'):
logging.warning('Currently, only the result from the \'BioCyc\' or \'KEGG+BioCyc\' runs can be directly added to a model.')
gap_analysis = rg.gapfill.gap_analysis(model_libsbml, config['gap_analysis_params'], filename)
logging.info(f'Gap analysis for {model_libsbml.getId()} with {config["gap_analysis_params"].get("db_to_compare")} was performed.')
logging.info(f'Complete Excel table is in file: {filename}.')
else:
gapfill = rg.gapfill.gapfill(model_libsbml, config['gap_analysis_params'], filename)
gap_analysis_stats = gapfill[0]
logging.info(f'Statistics on missing entites for {model_libsbml.getId()}:')
logging.info(gap_analysis_stats)
logging.info(f'Complete Excel table is in file: {filename}.')
model_libsbml = gapfill[-1]
logging.info(f'Gaps were filled in {model_libsbml.getId()}.')
elif config['gap_analysis']:
filename = f'{config["out_path"]}{name}_gap_analysis_{str(today)}.xlsx'
gap_analysis = rg.gapfill.gap_analysis(model_libsbml, config['gap_analysis_params'], filename)
logging.info(f'Gap analysis for {model_libsbml.getId()} with {config["gap_analysis_params"].get("db_to_compare")} was performed.')
if config["gap_analysis_params"].get("db_to_compare") != 'KEGG':
logging.info(f'Statistics on missing entites for {model_libsbml.getId()}:')
logging.info(gap_analysis[0])
logging.info(f'Complete Excel table is in file: {filename}.')
else:
gapfill = rg.gapfill.gapfill(model_libsbml, config['gap_analysis_params'], filename)
gap_analysis_stats = gapfill[0]
logging.info(f'Statistics on missing entites for {model_libsbml.getId()}:')
logging.info(gap_analysis_stats)
logging.info(f'Complete Excel table is in file: {filename}.')
model_libsbml = gapfill[-1]
elif config['gapfill_model']:
model_libsbml = rg.gapfill.gapfill_model(model_libsbml, config['gap_analysis_file'])
logging.info(f'Gaps were filled in {model_libsbml.getId()}.')
elif config['gap_analysis']:
filename = f'{config["out_path"]}{name}_gap_analysis_{str(today)}.xlsx'
gap_analysis = rg.gapfill.gap_analysis(model_libsbml, config['gap_analysis_params'], filename)
logging.info(f'Gap analysis for {model_libsbml.getId()} with {config["gap_analysis_params"].get("db_to_compare")} was performed.')
if config["gap_analysis_params"].get("db_to_compare") != 'KEGG':
logging.info(f'Statistics on missing entites for {model_libsbml.getId()}:')
logging.info(gap_analysis[0])
logging.info(f'Complete Excel table is in file: {filename}.')
elif config['gapfill_model']:
model_libsbml = rg.gapfill.gapfill_model(model_libsbml, config['gap_analysis_file'])
logging.info(f'Gaps were filled in {model_libsbml.getId()}.')

if (config['polish']):
model_libsbml = rg.polish.polish(model_libsbml, config['entrez_email'], config['id_db'], config['protein_fasta'], config['lab_strain'])
logging.info(model_libsbml.getId() + ' has been polished')

mods = [config['keggpathways'], config['sboterms'], config['charge_corr'], config['man_cur'], config['gapfill_model'], config['polish']]

if any(mods):
if config['model_out'] == 'stdout':
config['model_out'] = config['out_path'] + model_libsbml.getId() + '_modified_' + str(today) + '.xml'

rg.io.write_to_file(model_libsbml, config['model_out'])

if model_cobra is not None:
try:
model_cobra, errors = cobra.io.sbml.validate_sbml_model(config['model_out'])
logging.info(errors)
except (OSError):
model_cobra = None
logging.info('Model was invalidated during curation steps.')
if (config['polish']):
model_libsbml = rg.polish.polish(model_libsbml, config['entrez_email'], config['id_db'], config['protein_fasta'], config['lab_strain'])
logging.info(model_libsbml.getId() + ' has been polished')

mods = [config['keggpathways'], config['sboterms'], config['charge_corr'], config['man_cur'], config['gapfill_model'], config['polish']]

if any(mods):
if config['model_out'] == 'stdout':
config['model_out'] = config['out_path'] + model_libsbml.getId() + '_modified_' + str(today) + '.xml'

rg.io.write_to_file(model_libsbml, config['model_out'])

if model_cobra is not None:
try:
model_cobra, errors = cobra.io.sbml.validate_sbml_model(config['model_out'])
logging.info(errors)
except (OSError):
model_cobra = None
logging.info('Model was invalidated during curation steps.')

if (model_cobra != None):
logging.info(model_cobra.id + ' will be investigated.')
name, reac, metab, genes = rg.investigate.initial_analysis(model_libsbml)
orphans, deadends, disconnected = rg.investigate.get_orphans_deadends_disconnected(model_cobra)
mass_unbal, charge_unbal = rg.investigate.get_mass_charge_unbalanced(model_cobra)
egc = rg.investigate.get_egc(model_cobra)
if (config['visualize']):
logging.info('All visualizations can be found in the subfolder "visualization".')
sbo_fig = rg.investigate.plot_rea_sbo_single(model_libsbml).get_figure()
if (model_cobra != None):
logging.info(model_cobra.id + ' will be investigated.')
name, reac, metab, genes = rg.investigate.initial_analysis(model_libsbml)
orphans, deadends, disconnected = rg.investigate.get_orphans_deadends_disconnected(model_cobra)
mass_unbal, charge_unbal = rg.investigate.get_mass_charge_unbalanced(model_cobra)
egc = rg.investigate.get_egc(model_cobra)
if (config['visualize']):
logging.info('All visualizations can be found in the subfolder "visualization".')
sbo_fig = rg.investigate.plot_rea_sbo_single(model_libsbml).get_figure()

# saving the created visualizations
sbo_fig.savefig(config['out_path'] + 'visualization/' + str(model_cobra.id) + '_ReacPerSBO_' + str(today) + '.png', bbox_inches='tight')

# saving the created visualizations
sbo_fig.savefig(config['out_path'] + 'visualization/' + str(model_cobra.id) + '_ReacPerSBO_' + str(today) + '.png', bbox_inches='tight')

if (config['memote']):
score = rg.investigate.get_memote_score(rg.investigate.run_memote(model_cobra))
if (config['memote']):
score = rg.investigate.get_memote_score(rg.investigate.run_memote(model_cobra))
if(config['modelseed']):
charge_mismatch, formula_mismatch = rg.modelseed.compare_to_modelseed(model_cobra)

if(config['modelseed']):
charge_mismatch, formula_mismatch = rg.modelseed.compare_to_modelseed(model_cobra)

if (config['media'] != None):
growth_sim = rg.growth.get_growth_selected_media(model_cobra, config['media'], config['growth_basis'])

if (config['memote'] == True):
information = [[name], [reac], [metab], [genes], [score], orphans, deadends, disconnected, mass_unbal, charge_unbal]
model_params = pd.DataFrame(information, ['model name', '#reactions', '#metabolites', '#genes', 'memote score', 'orphans', 'deadends', 'disconnected', 'mass unbalanced', 'charge unbalanced']).T
if (config['media'] != None):
growth_sim = rg.growth.get_growth_selected_media(model_cobra, config['media'], config['growth_basis'])

if (config['memote'] == True):
information = [[name], [reac], [metab], [genes], [score], orphans, deadends, disconnected, mass_unbal, charge_unbal]
model_params = pd.DataFrame(information, ['model name', '#reactions', '#metabolites', '#genes', 'memote score', 'orphans', 'deadends', 'disconnected', 'mass unbalanced', 'charge unbalanced']).T
else:
information = [[name], [reac], [metab], [genes], orphans, deadends, disconnected, mass_unbal, charge_unbal]
model_params = pd.DataFrame(information, ['model name', '#reactions', '#metabolites', '#genes', 'orphans', 'deadends', 'disconnected', 'mass unbalanced', 'charge unbalanced']).T
with pd.ExcelWriter(config['out_path'] + name + '_' + str(today) + '.xlsx') as writer:
model_params.to_excel(writer, sheet_name='model params', index=False)
growth_sim.to_excel(writer, sheet_name='growth simulation', index=False)
egc.to_excel(writer, sheet_name='EGC test', index=False)
if(config['modelseed']):
charge_mismatch.to_excel(writer, sheet_name='charge mismatches', index=False)
formula_mismatch.to_excel(writer, sheet_name='formula mismatches', index=False)
logging.info('Single model growth simulation results are saved to ' + name + '_' + str(today) + '.xlsx')
else:
information = [[name], [reac], [metab], [genes], orphans, deadends, disconnected, mass_unbal, charge_unbal]
model_params = pd.DataFrame(information, ['model name', '#reactions', '#metabolites', '#genes', 'orphans', 'deadends', 'disconnected', 'mass unbalanced', 'charge unbalanced']).T
with pd.ExcelWriter(config['out_path'] + name + '_' + str(today) + '.xlsx') as writer:
model_params.to_excel(writer, sheet_name='model params', index=False)
growth_sim.to_excel(writer, sheet_name='growth simulation', index=False)
egc.to_excel(writer, sheet_name='EGC test', index=False)
if(config['modelseed']):
charge_mismatch.to_excel(writer, sheet_name='charge mismatches', index=False)
formula_mismatch.to_excel(writer, sheet_name='formula mismatches', index=False)
logging.info('Single model growth simulation results are saved to ' + name + '_' + str(today) + '.xlsx')
else:
logging.info('No valid model, investigation aborted!')
logging.info('No valid model, investigation aborted!')

if __name__ == "__main__":
try:
Expand Down
Loading

0 comments on commit 9b21b73

Please sign in to comment.