def plot_and_save_all_genes(data, fitter, fits, dirname,
show_change_distributions):
ensure_dir(dirname)
to_plot = []
genes = set(
) # use the genes from the fits and not from 'data' to support sharding (k_of_n)
for ds_fits in fits.itervalues():
for g, r in ds_fits.iterkeys():
genes.add(g)
for g in sorted(genes):
genedir = join(dirname, g[:g.index(cfg.exon_separator)]
) #does nothing if gene name does not contain the exon
ensure_dir(genedir)
filename = join(genedir, '{}.png'.format(g))
if isfile(filename):
print 'Figure already exists for gene {}. skipping...'.format(g)
continue
region_series_fits = _extract_gene_data(data, g, fits)
if show_change_distributions:
bin_centers = fits.change_distribution_params.bin_centers
else:
bin_centers = None
to_plot.append((g, region_series_fits, filename, bin_centers))
pool = Parallel(_plot_genes_job)
pool(pool.delay(*args) for args in to_plot)
def plot_and_save_all_exons_from_series(fits, exons_dir, series_dir):
to_plot = []
ensure_dir(exons_dir)
keys = set()
for ds_fits in fits.itervalues():
for g, r in ds_fits.iterkeys():
keys.add((g[:g.index(cfg.exon_separator)], r))
for g, r in keys:
gene_dir = join(exons_dir, g)
ensure_dir(gene_dir)
filename = join(gene_dir, '{}-{}.png'.format(g, r))
if isfile(filename):
print 'Figure already exists for gene {} in region {}. skipping...'.format(
g, r)
continue
to_plot.append((g, r, filename, series_dir))
pool = Parallel(_plot_exons_from_series_job)
pool(pool.delay(*args) for args in to_plot)
def plot_and_save_all_genes(data, fitter, fits, dirname, show_change_distributions):
ensure_dir(dirname)
to_plot = []
genes = set() # use the genes from the fits and not from 'data' to support sharding (k_of_n)
for ds_fits in fits.itervalues():
for g,r in ds_fits.iterkeys():
genes.add(g)
for g in sorted(genes):
filename = join(dirname, '{}.png'.format(g))
if isfile(filename):
print 'Figure already exists for gene {}. skipping...'.format(g)
continue
region_series_fits = _extract_gene_data(data,g,fits)
if show_change_distributions:
bin_centers = fits.change_distribution_params.bin_centers
else:
bin_centers = None
to_plot.append((g,region_series_fits,filename, bin_centers))
pool = Parallel(_plot_genes_job)
pool(pool.delay(*args) for args in to_plot)
def plot_and_save_all_series(data,
fitter,
fits,
dirname,
use_correlations,
show_change_distributions,
exons_layout=False,
figure_kw=None):
ensure_dir(dirname)
to_plot = []
for dsfits in fits.itervalues():
for (g, r), fit in dsfits.iteritems():
genedir = join(
dirname,
g[:g.index(cfg.exon_separator)] if exons_layout else g)
ensure_dir(genedir)
filename = join(genedir, 'fit-{}-{}.png'.format(g, r))
if isfile(filename):
print 'Figure already exists for {}@{}. skipping...'.format(
g, r)
continue
series = data.get_one_series(g, r)
if show_change_distributions and hasattr(
fit, 'change_distribution_weights'):
change_distribution = Bunch(
centers=fits.change_distribution_params.bin_centers,
weights=fit.change_distribution_weights,
)
else:
change_distribution = None
to_plot.append((series, fit, filename, use_correlations,
change_distribution, figure_kw))
if cfg.parallel_run_locally:
for args in to_plot:
_plot_series_job(*args)
else:
pool = Parallel(_plot_series_job)
pool(pool.delay(*args) for args in to_plot)
def plot_and_save_all_exons(data, fitter, fits, dirname):
ensure_dir(dirname)
to_plot = []
genes, regions = set(), set()
for ds_fits in fits.itervalues():
for g, r in ds_fits.iterkeys():
genes.add(g[:g.index(cfg.exon_separator)])
regions.add(r)
for g in sorted(genes):
for r in sorted(regions):
gene_dir = join(dirname, g)
ensure_dir(gene_dir)
filename = join(gene_dir, '{}-{}.png'.format(g, r))
if isfile(filename):
print 'Figure already exists for gene {} in region {}. skipping...'.format(
g, r)
continue
exons_series_fits = _extract_exons_data(data, g, r, fits)
if not np.count_nonzero(exons_series_fits):
continue
to_plot.append((g, r, exons_series_fits, filename))
pool = Parallel(_plot_exons_job)
pool(pool.delay(*args) for args in to_plot)
def plot_and_save_all_series(data, fitter, fits, dirname, use_correlations, show_change_distributions, figure_kw=None):
ensure_dir(dirname)
to_plot = []
for dsfits in fits.itervalues():
for (g,r),fit in dsfits.iteritems():
filename = join(dirname, 'fit-{}-{}.png'.format(g,r))
if isfile(filename):
print 'Figure already exists for {}@{}. skipping...'.format(g,r)
continue
series = data.get_one_series(g,r)
if show_change_distributions and hasattr(fit, 'change_distribution_weights'):
change_distribution = Bunch(
centers = fits.change_distribution_params.bin_centers,
weights = fit.change_distribution_weights,
)
else:
change_distribution = None
to_plot.append((series,fit,filename,use_correlations, change_distribution, figure_kw))
if cfg.parallel_run_locally:
for args in to_plot:
_plot_series_job(*args)
else:
pool = Parallel(_plot_series_job)
pool(pool.delay(*args) for args in to_plot)
def _make_layer(self, transitions, stride=1):
return Parallel(transitions)
def get_all_colantuoni_ages():
ids = range(749899, 750168) # copied from main accession page
pool = Parallel(get_age, verbosity=1)
ages = pool(pool.delay(_id) for _id in ids)
return ages
def get_all_colantuoni_ages():
ids = range(749899, 750168) # copied from main accession page
pool = Parallel(get_age, verbosity=1)
ages = pool(pool.delay(_id) for _id in ids)
return ages