long_label=hub_config['hub']['long_label'],
email=hub_config['hub']['email'],
genome=hub_config['hub']['genome'])
c = ChIPSeqConfig(
config, os.path.join(os.path.dirname(args.config),
'chipseq_patterns.yaml'))
# Set up subgroups based on unique values from columns specified in the config
df = pandas.read_csv(config['sampletable'], comment='#', sep='\t')
cols = hub_config['subgroups']['columns']
subgroups = []
for col in cols:
unique = list(df[col].unique())
mapping = {
sanitize(i, strict=True): sanitize(i, strict=False)
for i in unique
}
mapping['NA'] = 'NA'
s = SubGroupDefinition(
name=sanitize(col, strict=True),
label=col,
mapping=mapping,
)
subgroups.append(s)
# Add subgroups for algorithm and peaks-or-not for easier subsetting and
# sorting.
subgroups.append(
# SubGroupDefinitions later.
sample_subgroup_values = defaultdict(set)
bait_subgroup_values = defaultdict(set)
aggregates = {}
for comparison, vals, in config['4c']['comparisons'].items():
for kind in ['cis', 'nearbait']:
bait = vals['bait']
k = config['4c']['baits'][bait][kind + '_k']
for treatment in ['control', 'treatment', 'all']:
bait_subgroups = dict(
bait=sanitize(bait),
treatment=sanitize(treatment),
kind=sanitize(kind),
)
for key, val in bait_subgroups.items():
bait_subgroup_values[sanitize(key)].update(
[sanitize(val, False)])
if treatment == 'all':
local_fn = (
'4cker-output/{comparison}/{kind}_k{k}/'
'{bait}_{kind}_k{k}_adaptive_windows.bigbed'.format(
**locals()))
bait_bed_view.add_tracks(
Track(
hub_config = yaml.load(open(args.hub_config))
hub, genomes_file, genome, trackdb = default_hub(
hub_name=hub_config['hub']['name'],
short_label=hub_config['hub']['short_label'],
long_label=hub_config['hub']['long_label'],
email=hub_config['hub']['email'],
genome=hub_config['hub']['genome'])
# Set up subgroups based on the configured columns
df = pandas.read_table(config['sampletable'], comment='#')
cols = hub_config['subgroups']['columns']
subgroups = []
for col in cols:
unique = list(df[col].unique())
s = SubGroupDefinition(name=sanitize(col, strict=True),
label=col,
mapping={
sanitize(i, strict=True): sanitize(i,
strict=False)
for i in unique
})
subgroups.append(s)
# also add direction as a subgroup
# ASSUMPTION: stranded bigwigs were created
subgroups.append(
SubGroupDefinition(name='strand',
label='strand',
mapping={
'sense': 'sense',
hub, genomes_file, genome, trackdb = default_hub(
hub_name=hub_config['hub']['name'],
short_label=hub_config['hub']['short_label'],
long_label=hub_config['hub']['long_label'],
email=hub_config['hub']['email'],
genome=hub_config['hub']['genome']
)
# Set up subgroups based on the configured columns
df = pandas.read_csv(config['sampletable'], comment='#', sep='\t')
cols = hub_config['subgroups']['columns']
subgroups = []
for col in cols:
unique = list(df[col].unique())
s = SubGroupDefinition(
name=sanitize(col, strict=True),
label=col,
mapping={
sanitize(i, strict=True): sanitize(i, strict=False)
for i in unique}
)
subgroups.append(s)
# also add direction as a subgroup
# ASSUMPTION: stranded bigwigs were created
subgroups.append(
SubGroupDefinition(
name='strand',
label='strand',
mapping={'sense': 'sense', 'antisense': 'antisense'}))