def size_dist(inputs, paths_in, paths_out):
files = inputs['files']
path_figure = paths_out['path_figures']
plot_num = 0
sns.set_style("white")
plt.figure(figsize=(5, 3))
naming = ''
for fname in files:
naming += fname + '_'
path_analysis = paths_out['path_analysis'] + fname + '/readQC/'
data = ribo_util.unPickle(path_analysis + 'read_distribution')
df = ribo_util.dict_to_df(data, 'Length', 'fraction of total')
plt.plot(df, label=fname)
plt.title('Size Distribution')
plt.xlabel("Read Length")
plt.ylabel("Percent of Reads")
plt.legend(bbox_to_anchor=(1.05, 1), loc=2, borderaxespad=0.)
plt.savefig(path_figure + 'Comparison/size_dist' + '/sizedist_' + naming +
'.pdf',
dpi=400,
bbox_inches="tight")
plt.show()
for fname in files:
sns.set_style("white")
plt.figure(figsize=(5, 3))
path_analysis = paths_out['path_analysis'] + fname + '/readQC/'
data = ribo_util.unPickle(path_analysis + 'read_distribution')
df = ribo_util.dict_to_df(data, 'Length', 'fraction of total')
plt.plot(df, label=fname)
plt.title('Size Distribution')
plt.xlabel("Read Length")
plt.ylabel("Percent of Reads")
plt.legend(loc='upper right')
plt.savefig(path_figure + fname + '/sizedist.pdf',
dpi=400,
bbox_inches="tight")
size_plot_csv = pd.DataFrame(df)
size_plot_csv.to_csv(path_analysis + 'size_plot_values.csv')
plt.gcf().clear()
def plot_avggene(inputs, paths_in, paths_out, settings, settings_plot):
files = inputs['files']
shift = settings_plot['shift']
hmmax = settings_plot['HM_max']
ymax = settings_plot['ymax']
path_figure = paths_out['path_figures']
minlength = settings['minlength']
maxlength = settings['maxlength']
length_in_ORF = settings['length_in_ORF']
length_out_ORF = settings['length_out_ORF']
density_type = settings['density_type']
next_gene = settings['next_gene']
equal_weight = settings['equal_weight']
threshold = settings['threshold']
minlength_1 = str(minlength) + '_'
maxlength_1 = str(maxlength) + '_'
length_in_ORF_1 = str(length_in_ORF) + '_'
length_out_ORF_1 = str(length_out_ORF) + '_'
density_type_1 = density_type + '_'
next_gene_1 = str(next_gene) + '_'
equal_weight_1 = equal_weight + '_'
threshold_1 = str(threshold) + '_'
name_settings = length_in_ORF_1 + length_out_ORF_1 + next_gene_1 + threshold_1
name_settings += density_type_1 + equal_weight_1 + minlength_1 + maxlength_1
for fname in files:
plot_num = 0
sns.set_style("white")
plt.figure(figsize=(20, 5))
path_analysis = paths_out['path_analysis'] + fname + '/avggenes/'
data_start = ribo_util.unPickle(path_analysis + 'avg_start_' +
name_settings + '_all')
data_stop = ribo_util.unPickle(path_analysis + 'avg_stop_' +
name_settings + '_all')
data_startHM = ribo_util.unPickle(path_analysis + 'avg_start_' +
name_settings + '_HM')
data_stopHM = ribo_util.unPickle(path_analysis + 'avg_stop_' +
name_settings + '_HM')
xmax = len(data_start.keys())
data_start = ribo_util.dict_to_df(data_start, 'Position', 'Reads')
data_stop = ribo_util.dict_to_df(data_stop, 'Position', 'Reads')
data_startHM = ribo_util.heatmapdict_to_df(data_startHM, 'Length',
'Position', 'composition')
data_stopHM = ribo_util.heatmapdict_to_df(data_stopHM, 'Length',
'Position', 'composition')
data_startHM = data_startHM.reindex(index=data_startHM.index[::-1])
data_stopHM = data_stopHM.reindex(index=data_stopHM.index[::-1])
data_start.to_csv(path_analysis + 'start_all.csv')
data_stop.to_csv(path_analysis + 'stop_all.csv')
max_start = data_start["Reads"].max()
max_stop = data_stop["Reads"].max()
if ymax == 0:
if max_start > max_stop:
ymax = max_start
else:
ymax = max_stop
for graph in ['Start', 'Stop']:
plot_num += 1
if graph == 'Start':
data = data_start
elif graph == 'Stop':
data = data_stop
plt.subplot(2, 2, plot_num)
plt.plot(
data,
sns.xkcd_rgb["dark grey"],
)
plt.title(fname + ' ' + graph)
plt.ylabel("Reads")
plt.ylim(0, ymax)
plt.xlim(0, xmax)
sns.despine()
for graph in ['startHM', 'stopHM']:
plot_num += 1
if graph == 'startHM':
dataHM = data_startHM
elif graph == 'stopHM':
dataHM = data_stopHM
plt.subplot(2, 2, plot_num)
plot = sns.heatmap(dataHM,
cmap="ocean_r",
vmin=0,
vmax=hmmax,
cbar=False)
plt.setp(plot.get_xticklabels(), visible=False)
plt.setp(plot.get_xticklabels()[0::10], visible=True)
plt.setp(plot.get_yticklabels(), visible=False)
plt.setp(plot.get_yticklabels()[0::4], visible=True)
plt.savefig(path_figure + fname + '/avggene_' + name_settings + '.pdf',
dpi=400)
plt.show()
def plot_avggene_end(inputs, paths_in, paths_out, settings):
files = inputs['files']
shift = settings['shift']
hmmax = settings['HM_max']
for fname in files:
plot_num = 0
plt.figure(figsize=(20, 5))
path_analysis = paths_out['path_analysis'] + fname + '/'
data_start = ribo_util.unPickle(path_analysis + 'avg_start_all_end')
data_stop = ribo_util.unPickle(path_analysis + 'avg_stop_all_end')
data_startHM = ribo_util.unPickle(path_analysis + 'avg_start_HM_end')
data_stopHM = ribo_util.unPickle(path_analysis + 'avg_stop_HM_end')
xmax = len(data_start.keys())
data_start = ribo_util.dict_to_df(data_start, 'Position', 'Reads')
data_stop = ribo_util.dict_to_df(data_stop, 'Position', 'Reads')
data_startHM = ribo_util.heatmapdict_to_df(data_startHM, 'Length',
'Position', 'composition')
data_stopHM = ribo_util.heatmapdict_to_df(data_stopHM, 'Length',
'Position', 'composition')
data_start.to_csv(path_analysis + 'start_all.csv')
data_stop.to_csv(path_analysis + 'stop_all.csv')
max_start = data_start["Reads"].max()
max_stop = data_stop["Reads"].max()
if max_start > max_stop:
ymax = max_start
else:
ymax = max_stop
for graph in ['Start', 'Stop']:
plot_num += 1
if graph == 'Start':
data = data_start
elif graph == 'Stop':
data = data_stop
plt.subplot(2, 2, plot_num)
plt.plot(
data,
sns.xkcd_rgb["dark grey"],
)
plt.title(fname + ' ' + graph)
plt.ylabel("Reads")
plt.ylim(0, ymax)
plt.xlim(0, xmax)
sns.despine()
for graph in ['startHM', 'stopHM']:
plot_num += 1
if graph == 'startHM':
dataHM = data_startHM
elif graph == 'stopHM':
dataHM = data_stopHM
plt.subplot(2, 2, plot_num)
plot = sns.heatmap(dataHM,
cmap="ocean_r",
vmin=0,
vmax=hmmax,
cbar=False)
plt.setp(plot.get_xticklabels(), visible=False)
plt.setp(plot.get_xticklabels()[0::10], visible=True)
plt.setp(plot.get_yticklabels(), visible=False)
plt.setp(plot.get_yticklabels()[0::4], visible=True)
plt.show()