def venn_diagram_5prime(labels_a, labels_b, labels_c, figure_options=None):
# type: (Labels, Labels, Labels, FigureOptions) -> None
# first, reduce each set to common genes
list_labels_common_3prime = reduce_labels_to_genes_in_all(
[labels_a, labels_b, labels_c])
label_value_pair = numbers_for_3d_venn(*list_labels_common_3prime)
fig, ax = plt.subplots()
# venn3([set(get_set_gene_keys(labels)) for labels in list_labels_common_3prime],
# set_labels=[labels.name for labels in list_labels_common_3prime])
# create equal sized circles
v = venn3([1, 1, 1, 1, 1, 1, 1],
set_labels=[labels.name for labels in list_labels_common_3prime])
for key, value in label_value_pair.items():
v.get_label_by_id(key).set_text(value)
# Add title and annotation
FigureOptions.set_properties_for_axis(ax, figure_options)
if figure_options is not None and figure_options.save_fig is not None:
plt.savefig(figure_options.save_fig, bbox_inches='tight')
# Show it
plt.show()
def catplot(df, x, y, hue=None, kind="box", figure_options=None, **kwargs):
# type: (pd.DataFrame, str, str, str, Union[str, None], FigureOptions, Dict[str, Any]) -> None
sns_kwargs = get_value(kwargs, "sns_kwargs", dict())
g = sns.catplot(x=x, y=y, data=df, kind=kind, hue=hue, legend=False, aspect=1.5, **sns_kwargs)
if kind == "point":
plt.setp(g.ax.lines, linewidth=1) # set lw for all lines of g axes
# plt.setp(g.ax.lines, markersize=0) # set lw for all lines of g axes
#
# if fontsize:
# g.set_xlabels(x, fontsize=fontsize)
# g.set_ylabels(x, fontsize=fontsize)
FigureOptions.set_properties_for_axis(g.axes[0][0], figure_options)
legend = get_value(kwargs, "legend", "full")
legend_loc = get_value(kwargs, "legend_loc", None)
if hue is not None and legend:
title = get_value(kwargs, "legend_title", None)
if not legend_loc:
plt.legend(loc='center left', bbox_to_anchor=(1.05, 0.5), title=title)
else:
plt.legend(loc=legend_loc)
# plt.savefig(next_name(pd_work))
save_figure(figure_options)
plt.show()
def scatterplot(df, x, y, hue=None, figure_options=None, **kwargs):
# type: (pd.DataFrame, str, str, Union[str, None], FigureOptions, Dict[str, Any]) -> None
sns_kwargs = get_value(kwargs, "sns_kwargs", dict())
ax = get_value(kwargs, "ax", None)
identity = get_value(kwargs, "identity", False)
if not ax:
_, ax = plt.subplots()
g = sns.scatterplot(x=x, y=y, hue=hue, data=df, linewidth=0, **sns_kwargs)
if identity:
add_identity(ax, color="r", ls="--")
FigureOptions.set_properties_for_axis(ax, figure_options)
legend = get_value(kwargs, "legend", "full")
legend_loc = get_value(kwargs, "legend_loc", None)
if hue is not None and legend:
title = get_value(kwargs, "legend_title", None)
if not legend_loc:
plt.legend(loc='center left', bbox_to_anchor=(1.05, 0.5), title=title)
else:
plt.legend(loc=legend_loc)
save_figure(figure_options)
plt.show()
def plot_catplot(df, column_x, column_y, figure_options=None):
_, ax = plt.subplots()
sns.catplot(x=column_x, y=column_y, kind="bar", data=df)
FigureOptions.set_properties_for_axis(ax, figure_options)
if figure_options is not None and figure_options.save_fig is not None:
plt.savefig(figure_options.save_fig, bbox_index="tight")
plt.show()
def distplot(df, x, figure_options=None, **kwargs):
_, ax = plt.subplots()
sns_kwargs = get_value(kwargs, "sns_kwargs", dict())
if "kde" not in sns_kwargs:
sns_kwargs["kde"] = True
g = sns.distplot(df[x], bins=50, **sns_kwargs)
FigureOptions.set_properties_for_axis(g.axes, figure_options)
save_figure(figure_options)
plt.show()
def barplot(df, x, y, hue, figure_options=None, **kwargs):
sns_kwargs = get_value(kwargs, "sns_kwargs", dict())
ax = get_value(kwargs, "ax", None)
g = sns.barplot(x=x, y=y, data=df, hue=hue, ax=ax, **sns_kwargs)
if hue is not None:
plt.legend(loc='center left', bbox_to_anchor=(1.05, 0.5))
FigureOptions.set_properties_for_axis(g, figure_options)
plt.tight_layout()
save_figure(figure_options)
# plt.tight_layout(rect=[-0.3,0,1,1.2])
plt.show()
def kdeplot(df, x, y, hue=None, figure_options=None, **kwargs):
# type: (pd.DataFrame, str, str, Union[str, None], FigureOptions, Dict[str, Any]) -> None
sns_kwargs = get_value(kwargs, "sns_kwargs", dict())
_, ax = plt.subplots()
y_df = None if y is None else df[y]
g = sns.kdeplot(df[x], y_df, legend=False, **sns_kwargs)
if hue is not None:
plt.legend(loc='center left', bbox_to_anchor=(1.05, 0.5))
FigureOptions.set_properties_for_axis(ax, figure_options)
save_figure(figure_options)
plt.show()
def scatter(df, column_x, column_y, figure_options=None, **kwargs):
# type: (pd.DataFrame, str, str, FigureOptions, Dict[str, Any]) -> None
column_z = get_value(kwargs, "column_z", None)
identity = get_value(kwargs, "identity", False)
hue = df[column_z] if column_z is not None else None
_, ax = plt.subplots()
sns.jointplot(df[column_x], df[column_y], kind="scatter", alpha=0.3, s=10, linewidth=0)
#sns.scatterplot(df[column_x], df[column_y], hue=hue, alpha=0.3, s=10, linewidth=0)
if identity:
add_identity(ax, color="r", ls="--")
FigureOptions.set_properties_for_axis(ax, figure_options)
if figure_options is not None and figure_options.save_fig is not None:
plt.savefig(figure_options.save_fig, bbox_index="tight")
plt.show()
def plot_hist_by_group(df_data,
column_x,
column_group=None,
figure_options=None,
**kwargs):
# type: (pd.DataFrame, str, Union[str, None], FigureOptions, Dict[str, Any]) -> None
bins = get_value(kwargs, "bins", 10)
_, ax = plt.subplots()
cumulative = get_value(kwargs, "cumulative", False)
shade = False if cumulative else True
cut = [min(df_data[column_x]), max(df_data[column_x])]
if column_group is not None:
for name, df_group in df_data.groupby(column_group):
sns.distplot(df_group[column_x],
hist=False,
kde_kws={
"shade": shade,
"cumulative": cumulative
},
label=name)
else:
# sns.distplot(df_data[column_x], hist=True, kde_kws={"shade": shade, "cumulative": cumulative, "clip": cut})
sns.distplot(df_data[column_x],
bins=bins,
hist=True,
kde=False,
hist_kws={"edgecolor": "black"})
FigureOptions.set_properties_for_axis(ax, figure_options)
# plt.xlim([min(df_data[column_x]), max(df_data[column_x])])
if figure_options is not None and figure_options.save_fig is not None:
plt.savefig(figure_options.save_fig, bbox_index="tight")
plt.show()
def lmplot(df, x, y, hue=None, figure_options=None, **kwargs):
# type: (pd.DataFrame, str, str, Union[str, None], FigureOptions, Dict[str, Any]) -> None
sns_kwargs = get_value(kwargs, "sns_kwargs", dict())
if "aspect" not in sns_kwargs:
sns_kwargs["aspect"] = 2
g = sns.lmplot(x=x, y=y, hue=hue, data=df, legend=False, **sns_kwargs)
FigureOptions.set_properties_for_axis(g.axes[0][0], figure_options)
legend = get_value(kwargs, "legend", "full")
legend_loc = get_value(kwargs, "legend_loc", None)
if hue is not None and legend:
title = get_value(kwargs, "legend_title", None)
if not legend_loc:
g.axes[0][0].legend(loc='center left', bbox_to_anchor=(1.05, 0.5), title=title)
else:
g.axes[0][0].legend(loc=legend_loc)
save_figure(figure_options, fig=g.fig)
plt.subplots_adjust(right=1)
plt.show()
return g
def lineplot(df, x, y, hue=None, figure_options=None, **kwargs):
# type: (pd.DataFrame, str, str, Union[str, None], FigureOptions, Dict[str, Any]) -> None
sns_kwargs = get_value(kwargs, "sns_kwargs", dict())
ax = get_value(kwargs, "ax", None)
show = get_value(kwargs, "show", ax is None)
legend = get_value(kwargs, "legend", "full")
legend_loc = get_value(kwargs, "legend_loc", None)
legend_ncol = get_value(kwargs, "legend_ncol", 1)
identity = get_value(kwargs, "identity", False)
if not ax:
fig, ax = plt.subplots()
else:
fig = ax.get_figure()
g = sns.lineplot(x=x, y=y, hue=hue, data=df, ax=ax, legend=legend, **sns_kwargs)
if identity:
add_identity(ax, color="r", ls="--")
FigureOptions.set_properties_for_axis(ax, figure_options)
if hue is not None and legend:
title = get_value(kwargs, "legend_title", None)
if not legend_loc:
plt.legend(loc='center left', bbox_to_anchor=(1.05, 0.5), title=title, ncol=legend_ncol)
else:
plt.legend(loc=legend_loc, ncol=legend_ncol, title=title)
if title is not None and len(title) == 0:
handles, labels = ax.get_legend_handles_labels()
ax.legend(handles=handles[1:], labels=labels[1:], ncol=legend_ncol)
if show:
save_figure(figure_options, fig)
plt.show()
def main(env, args):
# type: (Environment, argparse.Namespace) -> None
df = pd.read_csv(args.pf_data)
df["chunk-size"] /= 1000
import matplotlib.pyplot as plt
fig, ax = plt.subplots()
sns.lineplot(df[df["Tool"] == "SBSP"], "chunk-size", "percentage-common-3prime-and-5prime-from-common-3prime",
hue="Genome",
sns_kwargs={"palette": CM.get_map("verified"), "linestyle": "dashed"},
ax=ax,
legend=False,
figure_options=FigureOptions(
xlabel="Chunk size (mb)",
ylabel="Accuracy",
ylim=[74, 101],
save_fig=next_name(env["pd-work"])
))
for l in ax.lines:
l.set_linestyle("--")
sns.lineplot(df[df["Tool"] == "GMS2"], "chunk-size", "percentage-common-3prime-and-5prime-from-common-3prime",
hue="Genome",
sns_kwargs={"palette": CM.get_map("verified")},
legend_loc="best",
legend_ncol=2,
ax=ax)
if args.with_mgm:
y_max = ax.get_ylim()[1]
ax.axvline(50, 0, y_max, color="grey", linestyle="dashed")
ax.axhline(74, 5, 49, color="grey", linestyle="dashed")
ax.annotate("MGM", (5, 72))
if "MGM" in set(df["Tool"]):
sns.lineplot(df[df["Tool"] == "MGM"], "chunk-size", "percentage-common-3prime-and-5prime-from-common-3prime",
hue="Genome",
sns_kwargs={"palette": CM.get_map("verified"), "linestyle": "-."},
ax=ax,
legend=False)
for l in ax.lines[len(ax.lines)-5:]:
l.set_linestyle(":")
fo = FigureOptions(
xlabel="Chunk size (mb)",
ylabel="Accuracy",
ylim=[74,101],
save_fig=next_name(env["pd-work"])
)
FigureOptions.set_properties_for_axis(ax, fo)
plt.savefig(fo.save_fig)
plt.show()