def test_open_wrong_database():
pycyc.open('_long_name_that_should_really_not_be_a_valid_pgdb')
will cause tests to fail even when the code is performing correctly,
but short of installing a purpose-built test PGDB I am not sure how to
avoid this.
"""
import pycyc
from nose.tools import raises, assert_is_instance, assert_raises
from pycyc.frame import Frame
# Specify the values which should be returned for true and false queries
# (which may change in the future)
_NIL = None
_T = True
ec = pycyc.open('ecoli')
sample_reaction = 'trypsyn-rxn' # '1.8.4.13-RXN', 'oropribtrans-rxn'
pathway_info = ('TRPSYN-PWY','RXN0-2381','OROPRIBTRANS-RXN')
(sample_pathway, pathway_reaction, pathway_non_reaction) = pathway_info
sample_compound = 'trp'
sample_gene = 'EG11024' # should encode at least one enzyme in at least one pwy
# should lie in at least one transcription unit
# and have at least one regulator
sample_enzyme = 'TRYPSYN' # should be of types chemical-change, small-molecule,
# non-transport
sample_protein = 'TRYPSYN'
sample_transport_reaction='TRANS-RXN-157'
# The sample transport protein should NOT be an enzyme of type non-transport
sample_transport_protein='CPLX-165'
sample_tf = 'PHOSPHO-CREB'
"""
Test interface object behavior not directly related to ptools interaction.
These tests require ecocyc and metacyc. As with the core Pathway Tools
API tests, they will fail if the specific objects used for testing
purposes change in certain ways due to database updates.
"""
import pycyc
ec1 = pycyc.open("ecoli")
ec2 = pycyc.open("ecoli")
meta = pycyc.open("meta")
def test_interface_to_str():
assert str(ec1) == "ecoli"
def test_interface_eq():
assert ec1 == ec2
assert not ec1 == meta
def test_interface_neq():
assert not ec1 != ec2
assert ec1 != meta
def main(args):
"""Run program
Args:
args (NameSpace): ArgParse arguments controlling program flow
"""
def shutdown(pid):
print('>>> Shutdown sequence initiated.')
print('>>> Terminating Pathway Tools LISP Daemon')
pid.terminate()
pid.wait()
print('>>> Daemon destroyed.')
print('>>> Until next time. :)')
print('>>> Hi, I am DPS (Derive Pathway Steps).')
print('>>> I will be analyzing pathways for you today.')
print('>>> I am using the {0} database as per your command.'
.format(args.database))
if args.database == 'metacyc':
# Obtain executable
pathway_tools = which('pathway-tools', path=args.executable)
if pathway_tools is None:
raise EnvironmentError('I cannot find pathway-tools: please '
'specify -e.')
else:
print('>>> I found pathway-tools: {0}.'.format(pathway_tools))
# Start pathway-tools daemon
while True:
print('>>> Summoning Pathway Tools LISP Daemon.')
pid = Popen([pathway_tools, '-lisp', '-api'],
stderr=open(os.devnull, 'w'),
stdout=open(os.devnull, 'w'))
print('>>> Let\'s give it five seconds to spawn.')
time.sleep(5)
if os.path.exists('/tmp/ptools-socket') and \
stat.S_ISSOCK(os.stat('/tmp/ptools-socket').st_mode):
print('>>> The daemon is is up!')
break
else:
print('>>> The daemon took too long to boot. :(')
print('>>> This makes me sad, so I will kill it.')
pid.kill()
print('>>> Let\'s wait five seconds for it to die!')
time.sleep(5)
pid.poll()
if pid.returncode is None:
raise CalledProcessError('Pathway Tools won\'t die!')
else:
print('>>> The daemon is dead!')
print('>>> I miss it. :( I\'m going to try again. :)')
atexit.register(shutdown, pid)
# Connect to daemon
try:
metacyc = pycyc.open('meta')
except IOError:
print('>>> I cannot connect to Pathway Tools Daemon.')
print('>>> Here is the original error message:')
raise
else:
print('>>> I have connected to the Pathway Tools Daemon.')
print('>>> Phenomenal cosmic powers! Itty bitty memory footprint!')
# Index genes file
print('>>> Indexing {0}.'.format(args.reactions_file.name))
reactions_to_genes = {}
start_time = time.time()
for line in args.reactions_file:
parts = line.strip().split()
reactions_to_genes[parts[0]] = (parts[1], parts[2:])
end_time = time.time()
print('>>> I indexed {0} reactions in {1} seconds.'
.format(str(len(reactions_to_genes)),
str(end_time - start_time)))
print('>>> I\'m so fast.')
# Index all pathways by name
print('>>> Time to index all the pathways from Metacyc.')
pathways = {}
start_time = time.time()
for frame in metacyc.all_pathways():
pathways[frame.common_name] = frame
end_time = time.time()
print('>>> I indexed {0} pathways in {1} seconds.'
.format(str(len(pathways)), str(end_time - start_time)))
print('>>> Aren\'t you proud of me?')
# Index gene abundance
print('>>> Recording gene abundances from {0}.'
.format(args.abundance_file.name))
abundances = {}
start_time = time.time()
for line in args.abundance_file:
gene, abundance = line.strip().split('\t')
abundances[gene] = abundance
end_time = time.time()
print('>>> I indexed {0} gene abundances in {1} seconds.'
.format(str(len(abundances)), str(end_time - start_time)))
# Obtain pathway of interest
print('>>> Time to do some science!')
print('>>> Note: you can input all or part of a pathway name.')
print('>>> Type "q" for input at any time to exit the program.')
while True: # Rest of program runs in a loop until user ends it
possibilities = {}
user_input = raw_input('>>> Enter a pathway: ')
if user_input.lower() == 'q':
break
for name, frame in pathways.items():
if user_input in name:
possibilities[name] = frame
if len(possibilities) == 0:
print('>>> I couldn\'t find any pathways matching your '
'request.')
print('>>> Try an alternative name for the pathway.')
continue
print('>>> I found {0} pathways matching your request.'
.format(str(len(possibilities))))
shutdown = False
restart = False
pathway = None
while True:
print('>>> Here are possible pathways:')
max_entry = len(possibilities) - 1
for possibility in enumerate(possibilities.items()):
print('{0}: {1}'.format(str(possibility[0]),
possibility[1][1].common_name))
path_num = raw_input('>>> Select a pathway ("r" to restart): ')
if path_num.lower() == 'q':
shutdown = True
break
elif path_num.lower() == 'r':
restart = True
break
else:
try:
path_num = int(path_num)
except ValueError:
print('>>> Your answer is not an integer.')
print('>>> I only understand integers.')
print('>>> Please correct.')
continue
if path_num > max_entry or path_num < 0:
print('>>> {0} is not a valid pathway.'
.format(str(path_num)))
print('>>> Valid pathways are: {0}.'.format(' '.join(
[str(i) for i in range(max_entry + 1)])))
print('>>> Try again.')
continue
pathway = possibilities[possibilities.keys()[path_num]]
print('>>> You selected: {0}.'.format(pathway.common_name))
print('>>> Neat! I\'ll analyze it now.')
break
if restart is True:
continue
if shutdown is True:
break
# Add genes and abundances to pathway reactions
print('>>> Collecting reactions in pathway.')
try:
if type(pathway.reaction_list) is list:
rxns = [str(rxn) for rxn in pathway.reaction_list]
else:
rxns = [str(pathway.reaction_list)]
except KeyError:
print('>>> I cannot access the reactions for this pathway. :(')
print('>>> I\'m sorry I\'ve failed you. :(')
print('>>> Please have me analyze something else.')
continue
print('>>> Analyzing pathway for key reactions.')
if hasattr(pathway, 'key_reactions') is True and\
pathway.key_reactions is not None:
key_rxns = [str(key) for key in pathway.key_reactions]
for rxn in enumerate(rxns):
if rxn[1] in key_rxns:
rxns[rxn[0]] = rxn[1] + '*'
print('>>> Acquiring gene families for each reaction from {0}.'
.format(args.reactions_file.name))
reactions = {}
for rxn in rxns:
rxn_name = re.sub('\*$', '', rxn)
if rxn_name in reactions_to_genes.keys():
ec, uniref_list = reactions_to_genes[rxn_name]
rxn_name = rxn + ' (' + ec + ')'
reactions[rxn_name] = {}
for uniref in uniref_list:
reactions[rxn_name][uniref] = 0.0
print('>>> Adding abundances from {0}.'
.format(args.abundance_file.name))
for rxn in reactions.keys():
for gene in reactions[rxn]:
if gene in abundances.keys():
reactions[rxn][gene] = abundances[gene]
print('>>> Removing unused gene families.')
for rxn in reactions.keys():
for uniref in reactions[rxn].keys():
if reactions[rxn][uniref] == 0.0:
del reactions[rxn][uniref]
for rxn in reactions.keys():
if reactions[rxn] == {}:
reactions[rxn] = 'None\tN/A'
continue
# Format reactions for printing
rxn_list = [pathway.common_name]
for rxn in reactions.keys():
if reactions[rxn] == 'None\tN/A':
temp = [rxn, ['None\tN/A']]
rxn_list.append(temp)
elif type(reactions[rxn]) is dict:
temp = [rxn]
for uniref in reactions[rxn].keys():
temp.append('{0}\t{1}'.format(uniref,
str(reactions[rxn][uniref])))
rxn_list.append(temp)
# Print output
print('>>> I\'ve finished analyzing everything!')
print('>>> Here it is (asterisks represent key reactions):')
rxn_print = [rxn for rxn in print_nested_list(rxn_list)]
for rxn in rxn_print:
print(rxn)
# Save output
print('>>> What file would you like me to save this to?')
print('>>> Type "n" if you don\'t want to save this output.')
while True:
out_file = raw_input('>>> File: ')
if out_file.lower() != 'n' and out_file.lower() != 'q':
try:
with open(out_file, 'w') as out_handle:
for rxn in rxn_print:
out_handle.write(rxn + os.linesep)
print('>>> Output written to {0}.'.format(out_file))
break
except IOError as error:
print('>>> I could not write to {0}.'.format(out_file))
print('>>> Original error:')
print(error)
print('>>> Let\'s try again (enter "n" to skip).')
elif out_file.lower() == 'q':
shutdown = True
break
else:
break
if shutdown is True:
break
print('>>> All done!')
print('>>> Let\'s do more science (enter "q" to exit program)!')
p_ignored_set = {} # dictionary of ignored pathways and blocking metabolites
r_ignored_set = set() # set of ignored reactions
m_generic_set = {} # dictionary of generic metabolites and how often they lead to a ignored reaction
p_generic_set = {} # dictionary of ignored pathways and how much ignored generic metabolites are responsible
r_generic = 0 # count generic reactions
r_total = 0 # count reactions
#
# I. input
#
print pycyc.all_orgs()
answer_org = raw_input("Which Organism shoulb be exportet to sbml? ")
org = pycyc.open(answer_org)
print "Loading", answer_org, "with", len(org.all_rxns(":all")), "reactions"
formula_dic = cyc.get_formula("./conf/formula.txt") # dictionary additional chemical formula for compounds
ec_dic = cyc.get_ec_dic("db/ec-names_brenda.txt") # dictionary with ec numbers and reaction names
answer_generic = raw_input(
"\nDo you want to take care of generic metabolites (e.g. tRNA, carboxylates)?\n All subclasses of a generic metabolite will be searched for instances (that is specific metabolites) and specific reactions (could be much more!) will be added instead of the generic one\n [y/n] "
)
care_generics = True if answer_generic == "y" else False
if care_generics:
answer_exceptions = raw_input(
"\nAre some metabolites (./conf/exceptions.txt) to be kept even if they are generic metabolites? This could be usefull e.g. to summarize lipid metabolism.\n [y/n] "
)
generic_exceptions = cyc.read_generic_exceptions("./conf/exceptions.txt") if answer_exceptions == "y" else []
else:
API tests, they will fail if the specific objects used for testing
purposes change in certain ways due to database updates.
"""
import re
import pycyc
from nose.tools import raises, assert_is_instance
from pycyc.frame import Frame
# This differs slightly from the attribute pattern defined in frame.py
# as here we are validating python attribute names, not finding LISP slot
# names which may be made pythonic
attribute_pattern = re.compile(r'[a-z_][a-z0-9_]*')
ec = pycyc.open('ecoli')
meta = pycyc.open('meta')
# Test frame equality
trp_ecoli_1 = Frame('trp',ec)
trp = trp_ecoli_1
trp_ecoli_2 = Frame('trp',ec)
trp_meta = Frame('trp',meta)
trp_single_value_slot = 'COMMON-NAME'
trp_many_value_slot = 'SYNONYMS'
trp_no_value_slot = 'SUPERATOMS'
trp_untranslatable_slot = 'SCHEMA?'
trp_empty_slot = 'SUPERATOMS'
class_ = Frame('|Genes|',ec)
instance = trp
class_member = 'EG11219'
