def plot_down_sampling(rna_file,
adt_file,
out_file,
probs=[i / 10.0 for i in range(9, 0, -1)],
n_threads=1,
dpi=500,
figsize=None):
data_gt = read_input(rna_file, mode='a')
adt_gt = read_input(adt_file, mode='a')
fracs, accuracy = down_sampling(data_gt,
adt_gt,
probs,
n_threads=n_threads)
plt.plot(fracs, accuracy, '.-')
ax = plt.gca()
ax.set_xlim(1.0, 0.0)
ax.set_ylim(0.79, 1.01)
vals = ax.get_yticks()
ax.set_yticklabels(['{:.0%}'.format(v) for v in vals])
ax.set_xlabel("Fraction of hashtag UMIs")
ax.set_ylabel("Consistency")
if figsize is not None:
plt.gcf().set_size_inches(*figsize)
plt.savefig(out_file, dpi=dpi)
plt.close()
def make_interactive_plots(input_file, plot_type, output_file, **kwargs):
adata = read_input(input_file, mode='r')
basis = transform_basis(plot_type)
if plot_type == 'diffmap' or plot_type == 'diffmap_pca':
df = pd.DataFrame(adata.obsm['X_{}'.format(plot_type)][:, 0:3],
index=adata.obs.index,
columns=[basis + i for i in ['1', '2', '3']])
if kwargs['isgene']:
coln = adata.var.index.get_loc(kwargs['attr'])
df.insert(0, 'Annotation', adata.X[:, coln].toarray().ravel())
else:
df.insert(0, 'Annotation', adata.obs[kwargs['attr']])
if not kwargs['isreal']:
iplot_library.scatter3d(df, output_file)
else:
iplot_library.scatter3d_real(df, output_file, kwargs['log10'])
else:
df = pd.DataFrame(adata.obsm['X_{}'.format(plot_type)],
index=adata.obs.index,
columns=[basis + i for i in ['1', '2']])
if kwargs['isgene']:
coln = adata.var.index.get_loc(kwargs['attr'])
df.insert(0, 'Annotation', adata.X[:, coln].toarray().ravel())
else:
df.insert(0, 'Annotation', adata.obs[kwargs['attr']])
if not kwargs['isreal']:
iplot_library.scatter(df, output_file)
else:
iplot_library.scatter_real(df, output_file, kwargs['log10'])
print(output_file + " is generated.")
adata.file.close()
def run_annotate_cluster(input_file,
output_file,
threshold,
ignoreNA=False,
json_file="human"):
start = time.time()
if json_file == "human_immune":
json_file = pkg_resources.resource_filename(
'scCloud.annotate_cluster', 'human_immune_cell_markers.json')
elif json_file == "mouse_immune":
json_file = pkg_resources.resource_filename(
'scCloud.annotate_cluster', 'mouse_immune_cell_markers.json')
elif json_file == "mouse_brain":
json_file = pkg_resources.resource_filename(
'scCloud.annotate_cluster', 'mouse_brain_cell_markers.json')
elif json_file == "human_brain":
json_file = pkg_resources.resource_filename(
'scCloud.annotate_cluster', 'human_brain_cell_markers.json')
data = read_input(input_file, mode='r')
with open(output_file, 'w') as fout:
annotate_cluster.annotate_clusters(data, json_file, threshold, fout,
ignoreNA)
data.file.close()
end = time.time()
print("Time spent for annotating clusters is {:.2f}s.".format(end - start))
def annotate_anndata_object(input_file, annotation):
data = read_input(input_file, mode='r+')
anno_attr, anno_str = annotation.split(':')
anno = anno_str.split(';')
data.obs[anno_attr] = [
anno[int(x) - 1] for x in data.obs[data.uns['de_labels']]
]
data.write(input_file)
def show_attributes(input_file, show_attributes, show_gene_attributes, show_values_for_attributes):
data = read_input(input_file, mode = 'r')
if show_attributes:
print("Available sample attributes in input dataset: {0}".format(', '.join(data.obs.columns.values)))
if show_gene_attributes:
print("Available gene attributes in input dataset: {0}".format(', '.join(data.var.columns.values)))
if not show_values_for_attributes is None:
for attr in show_values_for_attributes.split(','):
print("Available values for attribute {0}: {1}.".format(attr, ', '.join(np.unique(data.obs[attr]))))
def make_static_plots(input_file, plot_type, output_file, dpi=500, **kwargs):
adata = read_input(input_file, mode='r')
assert plot_type in pop_list
pop_set = pop_list[plot_type].copy()
for key, value in kwargs.items():
if value is None:
pop_set.add(key)
for key in pop_set:
kwargs.pop(key)
fig = getattr(plot_library, 'plot_' + plot_type)(adata, **kwargs)
fig.savefig(output_file, dpi=dpi)
print(output_file + " is generated.")
adata.file.close()