Merge pull request #94 from cdanielmachado/development

Development

carveme/cli/benchmark.py

@@ -16,6 +16,7 @@
     'sone': 'Shewanella oneidensis (MR-1)'
 }
 
+
 genomes = {
     'bsub': 'Bsubtilis_168.faa',
     'ecol': 'Ecoli_K12_MG1655.faa',

carveme/cli/build_universe.py

@@ -5,7 +5,8 @@
 import os
 import pandas as pd
 
-from carveme.universe.bigg_download import build_bigg_universe_model, download_model_specific_data, create_gpr_table
+from carveme.universe.bigg_download import build_bigg_universe_model, download_model_specific_data, create_gpr_table,\
+    download_gene_sequences
 from carveme.universe.curation import curate_universe
 from carveme.universe.thermodynamics import compute_bigg_gibbs_energy
 from carveme.reconstruction.utils import load_biomass_db
@@ -21,14 +22,17 @@ def maincall(mode, noheuristics=False, nothermo=False, allow_unbalanced=False, a
             universe_draft = outputfile
             model_specific_data = os.path.splitext(outputfile)[0] + '.csv'
             bigg_gprs = os.path.splitext(outputfile)[0] + '_gprs.csv'
+#            fasta_file = os.path.splitext(outputfile)[0] + '.faa'
         else:
             universe_draft = project_dir + config.get('generated', 'universe_draft')
             model_specific_data = project_dir + config.get('generated', 'model_specific_data')
             bigg_gprs = project_dir + config.get('generated', 'bigg_gprs')
+#            fasta_file = project_dir + config.get('input', 'fasta_file')
 
         build_bigg_universe_model(universe_draft)
         data = download_model_specific_data(model_specific_data)
-        create_gpr_table(data, outputfile=bigg_gprs)
+        gprs = create_gpr_table(data, outputfile=bigg_gprs)
+#        download_gene_sequences(gprs, outputfile=fasta_file)
 
     elif mode == 'thermo':
         universe_draft = project_dir + config.get('generated', 'universe_draft')

carveme/cli/carve.py

@@ -119,7 +119,7 @@ def maincall(inputfile, input_type='protein', outputfile=None, diamond_args=None
             universe_file = project_dir + config.get('generated', 'default_universe')
 
     try:
-        universe_model = load_cbmodel(universe_file, flavor=config.get('sbml', 'default_flavor'))
+        universe_model = load_cbmodel(universe_file, flavor='cobra')
         universe_model.id = model_id
     except IOError:
         available = '\n'.join(glob("{}{}universe_*.xml.gz".format(project_dir, config.get('generated', 'folder'))))

carveme/config.cfg

@@ -1,13 +1,14 @@
 [input]
 equilibrator_compounds = data/input/equilibrator_compounds.tsv.gz
-diamond_db = data/input/bigg_proteins.dmnd
 biomass_library = data/input/biomass_db.tsv
 media_library = data/input/media_db.tsv
 bigg_models = data/input/bigg_models.csv
 metabolomics = data/input/metabolomics_park2016.csv
 unbalanced_metabolites = data/input/unbalanced_metabolites.csv
 manually_curated = data/input/manually_curated.csv
 refseq = data/input/refseq_release_201.tsv.gz
+fasta_file = data/input/bigg_proteins.faa
+diamond_db = data/input/bigg_proteins.dmnd
 
 [generated]
 folder = data/generated/
@@ -17,6 +18,7 @@ universe_draft = data/generated/universe_draft.xml.gz
 default_universe = data/generated/universe_bacteria.xml.gz
 bigg_gibbs = data/generated/bigg_gibbs.csv
 
+
 [solver]
 default_solver = cplex
 feas_tol = 1e-9
@@ -32,7 +34,7 @@ concentration_max = 1e-2
 measured_fold_change = 10
 
 [sbml]
-default_flavor = cobra
+default_flavor = fbc2
 
 [gapfill]
 max_uptake = 100

carveme/universe/bigg_download.py

@@ -16,7 +16,7 @@
 
 def progress(i, n):
     p = int((i+1)*100.0 / n)
-    sys.stdout.write("\r{}%".format(p))
+    sys.stdout.write(f"\r{p}%")
     sys.stdout.flush()
 
 
@@ -75,13 +75,13 @@ def build_reaction(model, bigg_id, ignore_pseudo_reactions=True, only_pseudo_rea
 
     for entry in data['metabolites']:
         comp = str(entry['compartment_bigg_id'])
-        c_id = 'C_{}'.format(comp)
+        c_id = f'C_{comp}'
 
         if c_id not in model.compartments:
             build_compartment(model, comp)
 
         met = entry['bigg_id']
-        m_id = 'M_{}_{}'.format(met, comp)
+        m_id = f'M_{met}_{comp}'
 
         if m_id not in model.metabolites:
             build_metabolite(model, met, comp)
@@ -119,11 +119,11 @@ def build_metabolite(model, bigg_id, compartment):
 
     """
     data = get_request(metabolites_url + bigg_id)
-    m_id = 'M_{}_{}'.format(bigg_id, compartment)
+    m_id = f'M_{bigg_id}_{compartment}'
     c_id = 'C_' + compartment
     name = str(data['name'])
     metabolite = Metabolite(m_id, name, c_id)
-    model.add_metabolite(metabolite, clear_tmp=False)
+    model.add_metabolite(metabolite)
     extract_metabolite_metadata(metabolite, data, compartment)
 
 
@@ -230,7 +230,7 @@ def download_model_specific_data(outputfile=None):
 
     for i, entry in enumerate(data['results']):
         model_id = entry['bigg_id']
-        model = get_request('{}{}/download'.format(models_url, model_id))
+        model = get_request(f'{models_url}{model_id}/download')
 
         for reaction in model['reactions']:
             r_id = reaction['id']
@@ -304,3 +304,17 @@ def create_gpr_table(model_specific_data, reactions=None, outputfile=None):
         df.to_csv(outputfile, index=False)
 
     return df
+
+
+def download_gene_sequences(gprs, outputfile):
+    df = gprs[['gene', 'model']].drop_duplicates()
+    data = {}
+    for _, row in df.iterrows():
+        entry = get_request(f'{models_url}/{row["model"]}/genes/{row["gene"]}')
+        data[(row['gene'], row['model'])] = entry['protein_sequence']
+
+    with open(outputfile, 'w') as f:
+        for (gene, model), seq in data.items():
+            f.write(f'>{model}.{gene}\n')
+            f.write(seq + '\n')
+

carveme/universe/curation.py

@@ -1,11 +1,11 @@
 import numpy as np
-from reframed.model.transformation import disconnected_metabolites
+from reframed.core.transformation import disconnected_metabolites
 from carveme.reconstruction.utils import create_exchange_reactions, add_biomass_equation, create_sink_reactions, \
     add_maintenance_atp
 from reframed import simplify
 from reframed import save_cbmodel
 from carveme.universe.thermodynamics import compute_flux_bounds
-from reframed.experimental.elements import parse_formula
+from reframed.core.elements import parse_formula
 
 
 def elemental_balance(formulas, tol=1e-6):

setup.py

@@ -9,7 +9,7 @@
     readme = readme_file.read()
 
 requirements = [
-    "reframed>=1.1",
+    "reframed>=1.2",
     "pandas>=0.20.0",
     "requests>=2.18"
 ]
@@ -23,7 +23,18 @@
         'data/input/media_db.tsv',
         'data/input/metabolomics_park2016.csv',
         'data/input/unbalanced_metabolites.csv',
-        'data/generated/bigg_gibbs.csv'
+        'data/input/bigg_proteins.faa',
+        'data/input/bigg_proteins.dmnd',
+        'data/input/equilibrator_compounds.tsv.gz',
+        'data/input/refseq_release_201.tsv.gz',
+        'data/generated/bigg_gibbs.csv',
+        'data/generated/bigg_gprs.csv.gz',
+        'data/generated/model_specific_data.csv.gz',
+        'data/generated/universe_draft.xml.gz',
+        'data/generated/universe_bacteria.xml.gz',
+        'data/generated/universe_grampos.xml.gz',
+        'data/generated/universe_gramneg.xml.gz',
+        'data/generated/universe_archaea.xml.gz',
     ]
 }
 
@@ -33,20 +44,19 @@
     version='1.3.0',
     description="CarveMe: automated metabolic model reconstruction",
     long_description=readme,
-    author="Daniel Machado, Sergej Andrejev",
+    author="Daniel Machado",
     author_email='cdanielmachado@gmail.com',
     url='https://github.com/cdanielmachado/carveme',
     entry_points={
         'console_scripts': [
             'build_universe=carveme.cli.build_universe:main',
             'carve=carveme.cli.carve:main',
-            'carveme_init=carveme.cli.carveme_init:main',
             'gapfill=carveme.cli.gapfill:main',
             'merge_community=carveme.cli.merge_community:main',
             'benchmark=carveme.cli.benchmark:main',
         ],
     },
-#    package_dir={'':'src'},
+    #   package_dir={'':'src'},
     packages=find_packages(),
     include_package_data=True,
     package_data=included_files,
@@ -64,6 +74,6 @@
         'License :: OSI Approved :: Apache Software License',
     ],
     setup_requires=['setuptools_scm']
-#    test_suite='tests',
-#    tests_require=test_requirements,
+    #    test_suite='tests',
+    #    tests_require=test_requirements,
 )