workdir = '/'.join(contig_input_file.split('/')[:-1])

contig_matrix = pd.read_csv(contig_input_file,sep='\t',index_col=0)

x_labels = np.array(contig_matrix.index)

num_x_labels = range(len(x_labels))

#______________________________Contig Data_____________________________________

# Read the contig data

data_1_contig_present = np.matrix(contig_matrix).T

data_1_y_labels = contig_matrix.columns

# replace substring in sample name

data_1_y_labels = np.core.defchararray.replace(np.array(data_1_y_labels,dtype=str), 'sample_', 'contigs ')

# print a text file with the loci indeces and the corresponding loci names

new_locus_list = x_labels

locus_index_overview = pd.DataFrame({'loci':new_locus_list})

locus_index_overview.to_csv(os.path.join(workdir,'locus_index_overview.txt'),sep='\t',header=False)

#___________________________Plotting settings___________________________________

height,width = data_1_contig_present.shape

fig = plt.figure(figsize=(20,8))

#fig.subplots_adjust(top=1, bottom=0.0, left=0.2, right=0.99)

for i,m in enumerate(data_1_contig_present):

ax = plt.subplot(height, 1, i+1)

ax.tick_params(left='off',bottom='off',labelleft='off')

# Only plot x-axis for last row

if not i == height-1:

ax.xaxis.set_major_formatter(plt.NullFormatter())

#plt.axis("off")

if data_1_y_labels[i] == 'contig alignment':

plt.imshow(data_1_contig_present[i], aspect='auto', cmap='binary', origin='lower')

else:

plt.imshow(data_1_contig_present[i], aspect='auto', cmap='GnBu', origin='lower')

pos = list(ax.get_position().bounds)

fig.text(pos[0] - 0.01, pos[1], data_1_y_labels[i], horizontalalignment='right')

plt.xlabel('exon index')

#plt.colorbar()

workdir = '/'.join(contig_input_file.split('/')[:-1])

contig_matrix = pd.read_csv(contig_input_file,sep='\t',index_col=0)

x_labels = np.array(contig_matrix.index)

num_x_labels = range(len(x_labels))

#______________________________Contig Data_____________________________________

# Read the contig data

data_1_contig_present = np.matrix(contig_matrix).T

data_1_y_labels = contig_matrix.columns

# replace substring in sample name

data_1_y_labels = np.core.defchararray.replace(np.array(data_1_y_labels,dtype=str), 'sample_', 'contigs ')

#_______________________________Contig Alignment Data__________________________

# Get the alignment files and make list of loci with alignments

alignment_files = glob.glob(os.path.join(alignment_folder, '*.fasta'))

list_of_loci_with_alignments = [re.sub('.fasta','',al.split('/')[-1]) for al in alignment_files]

# Create 1-dimensional matrix and fill with info which loci have alignment data

presence_absence_df = pd.DataFrame({'loci':x_labels,'presence':0})

for locus in list_of_loci_with_alignments:

row_index = presence_absence_df[presence_absence_df.loci == locus].index

presence_absence_df.loc[row_index,'presence'] = 1

data_2_contig_alignment = np.matrix(presence_absence_df.presence)

data_2_y_labels = np.array('contig alignment')

#_________________________Combine contig and alignment data_____________________

contig_data_subset = np.vstack([data_1_contig_present, data_2_contig_alignment])

y_labels_contig_data = np.append(data_1_y_labels,data_2_y_labels)

# print a text file with the loci indeces and the corresponding loci names

new_locus_list = x_labels

locus_index_overview = pd.DataFrame({'loci':new_locus_list})

locus_index_overview.to_csv(os.path.join(workdir,'locus_index_overview.txt'),sep='\t',header=False)

#___________________________Plotting settings___________________________________

height,width = contig_data_subset.shape

fig = plt.figure(figsize=(20,8))

#fig.subplots_adjust(top=1, bottom=0.0, left=0.2, right=0.99)

for i,m in enumerate(contig_data_subset):

ax = plt.subplot(height, 1, i+1)

ax.tick_params(left='off',bottom='off',labelleft='off')

# Only plot x-axis for last row

if not i == height-1:

ax.xaxis.set_major_formatter(plt.NullFormatter())

#plt.axis("off")

if y_labels_contig_data[i] == 'contig alignment':

plt.imshow(contig_data_subset[i], aspect='auto', cmap='binary', origin='lower')

else:

plt.imshow(contig_data_subset[i], aspect='auto', cmap='GnBu', origin='lower')

pos = list(ax.get_position().bounds)

fig.text(pos[0] - 0.01, pos[1], y_labels_contig_data[i], horizontalalignment='right')

plt.xlabel('exon index')

#plt.colorbar()

mpl.rcParams.update({'font.size': font_size})

workdir = '/'.join(read_cov_file.split('/')[:-1])

contig_matrix = pd.read_csv(contig_input_file,sep='\t',index_col=0)

x_labels = np.array(contig_matrix.index)

num_x_labels = range(len(x_labels))

#______________________________1. Contig Data_____________________________________

# Read the contig data

data_1_contig_present = np.matrix(contig_matrix).T

data_1_y_labels = contig_matrix.columns

# replace substring in sample name

data_1_y_labels = np.core.defchararray.replace(np.array(data_1_y_labels,dtype=str), string_to_remove_from_sample_names, 'contigs ')

#_______________________________2. Contig Alignment Data__________________________

# Get the alignment files and make list of loci with alignments

alignment_files = glob.glob(os.path.join(alignment_folder, '*.fasta'))

list_of_loci_with_alignments = [re.sub('.fasta','',al.split('/')[-1]) for al in alignment_files]

# Create 1-dimensional matrix and fill with info which loci have alignment data

presence_absence_df = pd.DataFrame({'loci':x_labels,'presence':0})

for locus in list_of_loci_with_alignments:

row_index = presence_absence_df[presence_absence_df.loci == locus].index

presence_absence_df.loc[row_index,'presence'] = 1

data_2_contig_alignment = np.matrix(presence_absence_df.presence)

data_2_y_labels = np.array('contig alignment')

#_______________________________3. Reference-assembly Data__________________________

# Get the data as pandas dataframe

unsorted_read_cov_data = pd.read_csv(read_cov_file, sep = '\t',index_col=0)

locus_selection=False

if 'sum_per_locus' in unsorted_read_cov_data.columns:

unsorted_read_cov_data = unsorted_read_cov_data.iloc[:,:-1]

locus_selection=True

# sort columns in df

temp_read_cov_data = unsorted_read_cov_data[sorted(unsorted_read_cov_data.columns)].sort_index()

# add row of 0's for all missing loci

loci_in_df = list(temp_read_cov_data.index)

for locus in list(x_labels):

if locus not in loci_in_df:

temp_read_cov_data.loc[locus] = [0.0]*len(temp_read_cov_data.columns)

# sort by index again

read_cov_data = temp_read_cov_data.sort_index()

# turn df into matrix

data_3_read_cov = np.matrix(read_cov_data).T

# lets use the same labels as for the contig data

data_3_y_labels = np.core.defchararray.replace(data_1_y_labels, 'contigs', 'coverage')

#___________________________Combine all Data___________________________________

combined_data = np.vstack([data_1_contig_present, data_2_contig_alignment,data_3_read_cov])

tmp_combined_y_labels = np.append(data_1_y_labels,data_2_y_labels)

combined_y_labels = np.append(tmp_combined_y_labels,data_3_y_labels)

height,width = combined_data.shape

# Define the range for the heatmap, above the maximum everything will be colored in the highest color

norm=None

if norm_value:

norm = mpl.colors.Normalize(vmin=0, vmax=norm_value)

if locus_selection and reduce:

print('Reducing final matrix to selected loci.')

# Reduce the data matrix to only those columns for which we have extracted values in the selected loci

boolean = combined_data[-1]>0

reduced_data = np.matrix([np.matrix.tolist(ind[boolean])[0] for ind in combined_data])

# also print a text file with the loci indeces and the corresponding loci names

new_locus_list = x_labels[np.array(boolean)[0]]

locus_index_overview = pd.DataFrame({'loci':new_locus_list})

locus_index_overview.to_csv(os.path.join(workdir,'locus_index_overview.txt'),sep='\t',header=False)

# continue with plotting

fig = plt.figure(figsize=(13.5,8))

for i,m in enumerate(reduced_data):

ax = plt.subplot(height, 1, i+1)

ax.tick_params(left='off',bottom='off',labelleft='off')

# Only plot x-axis for last row

if not i == height-1:

ax.xaxis.set_major_formatter(plt.NullFormatter())

#plt.axis("off")

if combined_y_labels[i] == 'contig alignment':

plt.imshow(reduced_data[i], aspect='auto', cmap='Greens_r', origin='lower')

elif 'contigs' in combined_y_labels[i]:

plt.imshow(reduced_data[i], aspect='auto', cmap='GnBu', origin='lower')

else:

plt.imshow(reduced_data[i], aspect='auto', cmap='hot_r',norm=norm, origin='lower')#,clim=(0.0, 10))

pos = list(ax.get_position().bounds)

fig.text(pos[0] - 0.01, pos[1], combined_y_labels[i], horizontalalignment='right')

plt.xlabel('exon index')

else:

# print a text file with the loci indeces and the corresponding loci names

new_locus_list = x_labels

locus_index_overview = pd.DataFrame({'loci':new_locus_list})

locus_index_overview.to_csv(os.path.join(workdir,'locus_index_overview.txt'),sep='\t',header=False)

if not number_of_rows:

#_______________________________Plot Combined Data_____________________________

fig = plt.figure(figsize=(20,8))

#fig.subplots_adjust(top=1, bottom=0.0, left=0.2, right=0.99)

for i,m in enumerate(combined_data):

ax = plt.subplot(height, 1, i+1)

ax.tick_params(left='off',bottom='off',labelleft='off')

# Only plot x-axis for last row

if not i == height-1:

ax.xaxis.set_major_formatter(plt.NullFormatter())

#plt.axis("off")

if combined_y_labels[i] == 'contig alignment':

plt.imshow(combined_data[i], aspect='auto', cmap='Greens', origin='lower')

elif 'contigs' in combined_y_labels[i]:

plt.imshow(combined_data[i], aspect='auto', cmap='GnBu', origin='lower')

else:

plt.imshow(combined_data[i], aspect='auto', cmap='hot_r',norm=norm, origin='lower')#,clim=(0.0, 10))

pos = list(ax.get_position().bounds)

fig.text(pos[0] - 0.01, pos[1], combined_y_labels[i], horizontalalignment='right')

plt.xlabel('exon index')

#plt.colorbar()

elif number_of_rows:

images = []

#_______________________________Plot Split Data_____________________________

# Split dataset for better readability

columns_per_row = int(combined_data.shape[1]/number_of_rows)

remainder = combined_data.shape[1]%number_of_rows

# start and end of first row, adding the remainder to the first row

b = 0

n = columns_per_row+remainder

subset_dict = {}

# iterate through chunks

for i in range(number_of_rows):

data_chunk = combined_data[:,b:n]

subset_dict.setdefault('%i,%i' %(b,n),data_chunk)

b=n

n+=columns_per_row

for j in subset_dict.keys():

split_data = subset_dict[j]

data_range = j.split(',')

num_x_labels = np.arange(int(data_range[0]),int(data_range[-1]))

fig = plt.figure(figsize=(20,8))

#fig.subplots_adjust(top=1, bottom=0.0, left=0.2, right=0.99)

for i,m in enumerate(split_data):

ax = plt.subplot(height, 1, i+1)

ax.tick_params(left='off',bottom='off',labelleft='off')

# Only plot x-axis for last row

if not i == height-1:

ax.xaxis.set_major_formatter(plt.NullFormatter())

#plt.axis("off")

if combined_y_labels[i] == 'contig alignment':

plt.imshow(split_data[i], aspect='auto', cmap='Greens', origin='lower')

elif 'contigs' in combined_y_labels[i]:

plt.imshow(split_data[i], aspect='auto', cmap='GnBu', origin='lower')

else:

plt.imshow(split_data[i], aspect='auto', cmap='hot_r',norm=norm, origin='lower')#,clim=(0.0, 10))

pos = list(ax.get_position().bounds)

fig.text(pos[0] - 0.01, pos[1], combined_y_labels[i], horizontalalignment='right')

# make sure to have 10 ticks on the x-axis (ensured by dividing the total lenght by 9 and using the resulting value as stepsize)

tick_step_size = split_data[i].shape[1]/9

# get the desired indeces of the x-values that shall carry ticks on x-axis

xi = np.arange(0, split_data[i].shape[1],int(tick_step_size))

# get the corresponding x-values from the num_x_labels variable (dicitonary keys)

x = np.arange(num_x_labels[0], num_x_labels[-1], int(tick_step_size))

plt.xticks(xi,x)

plt.xlabel('exon index')

fig.savefig(os.path.join(workdir,'contig_exon_coverage_matrix_%s.png'%j), dpi = 500)

images.append(fig)

if not number_of_rows:

fig.savefig(os.path.join(outdir,'contig_yield_and_msas_and_readcov_overview.png'),bbox_inches='tight', dpi = 500)

else:

workdir = '/'.join(contig_input_file.split('/')[:-1])

contig_matrix = pd.read_csv(contig_input_file,sep='\t',index_col=0)

x_labels = np.array(contig_matrix.index)

num_x_labels = range(len(x_labels))

#______________________________Contig Data_____________________________________

# Read the contig data

data_1_contig_present = np.matrix(contig_matrix).T

data_1_y_labels = contig_matrix.columns

# replace substring in sample name

data_1_y_labels = np.core.defchararray.replace(np.array(data_1_y_labels,dtype=str), 'sample_', 'contigs ')

# print a text file with the loci indeces and the corresponding loci names

new_locus_list = x_labels

locus_index_overview = pd.DataFrame({'loci':new_locus_list})

locus_index_overview.to_csv(os.path.join(workdir,'locus_index_overview.txt'),sep='\t',header=False)

#___________________________Plotting settings___________________________________

height,width = data_1_contig_present.shape

fig = plt.figure(figsize=(20,8))

#fig.subplots_adjust(top=1, bottom=0.0, left=0.2, right=0.99)

for i,m in enumerate(data_1_contig_present):

ax = plt.subplot(height, 1, i+1)

ax.tick_params(left='off',bottom='off',labelleft='off')

# Only plot x-axis for last row

if not i == height-1:

ax.xaxis.set_major_formatter(plt.NullFormatter())

#plt.axis("off")

if data_1_y_labels[i] == 'contig alignment':

plt.imshow(data_1_contig_present[i], aspect='auto', cmap='binary', origin='lower')

else:

plt.imshow(data_1_contig_present[i], aspect='auto', cmap='GnBu', origin='lower')

pos = list(ax.get_position().bounds)

fig.text(pos[0] - 0.01, pos[1], data_1_y_labels[i], horizontalalignment='right')

plt.xlabel('exon index')

#plt.colorbar()

workdir = '/'.join(contig_input_file.split('/')[:-1])

contig_matrix = pd.read_csv(contig_input_file,sep='\t',index_col=0)

x_labels = np.array(contig_matrix.index)

num_x_labels = range(len(x_labels))

#______________________________Contig Data_____________________________________

# Read the contig data

data_1_contig_present = np.matrix(contig_matrix).T

data_1_y_labels = contig_matrix.columns

# replace substring in sample name

data_1_y_labels = np.core.defchararray.replace(np.array(data_1_y_labels,dtype=str), 'sample_', 'contigs ')

#_______________________________Contig Alignment Data__________________________

# Get the alignment files and make list of loci with alignments

alignment_files = glob.glob(os.path.join(alignment_folder, '*.fasta'))

list_of_loci_with_alignments = np.array([re.sub('.fasta','',al.split('/')[-1]) for al in alignment_files]).astype(int)

# Create 1-dimensional matrix and fill with info which loci have alignment data

presence_absence_df = pd.DataFrame({'loci':x_labels,'presence':0})

for locus in list_of_loci_with_alignments:

row_index = presence_absence_df[presence_absence_df.loci == locus].index

presence_absence_df.loc[row_index,'presence'] = 1

data_2_contig_alignment = np.array(presence_absence_df.presence)

data_2_y_labels = np.array('contig alignment')

#_________________________Combine contig and alignment data_____________________

contig_data_subset = np.vstack([data_1_contig_present, data_2_contig_alignment])

y_labels_contig_data = np.append(data_1_y_labels,data_2_y_labels)

# print a text file with the loci indeces and the corresponding loci names

new_locus_list = x_labels

locus_index_overview = pd.DataFrame({'loci':new_locus_list})

locus_index_overview.to_csv(os.path.join(workdir,'locus_index_overview.txt'),sep='\t',header=False)

#___________________________Plotting settings___________________________________

height,width = contig_data_subset.shape

fig = plt.figure(figsize=(20,8))

#fig.subplots_adjust(top=1, bottom=0.0, left=0.2, right=0.99)

for i,m in enumerate(contig_data_subset):

ax = plt.subplot(height, 1, i+1)

ax.tick_params(left='off',bottom='off',labelleft='off')

# Only plot x-axis for last row

if not i == height-1:

ax.xaxis.set_major_formatter(plt.NullFormatter())

#plt.axis("off")

if y_labels_contig_data[i] == 'contig alignment':

plt.imshow(contig_data_subset[i], aspect='auto', cmap='binary', origin='lower')

else:

plt.imshow(contig_data_subset[i], aspect='auto', cmap='GnBu', origin='lower')

pos = list(ax.get_position().bounds)

fig.text(pos[0] - 0.01, pos[1], y_labels_contig_data[i], horizontalalignment='right')

plt.xlabel('exon index')

#plt.colorbar()

mpl.rcParams.update({'font.size': font_size})

workdir = '/'.join(read_cov_file.split('/')[:-1])

contig_matrix = pd.read_csv(contig_input_file,sep='\t',index_col=0)

x_labels = np.array(contig_matrix.index)

num_x_labels = range(len(x_labels))

#______________________________1. Contig Data_____________________________________

# Read the contig data

data_1_contig_present = np.array(contig_matrix).T

data_1_y_labels = contig_matrix.columns

# replace substring in sample name

data_1_y_labels = np.array([i.replace(string_to_remove_from_sample_names,'')+'_contigs' for i in data_1_y_labels])

#_______________________________2. Contig Alignment Data__________________________

# Get the alignment files and make list of loci with alignments

alignment_files = glob.glob(os.path.join(alignment_folder, '*.fasta'))

list_of_loci_with_alignments = np.array([re.sub('.fasta','',al.split('/')[-1]) for al in alignment_files]).astype(int)

# Create 1-dimensional matrix and fill with info which loci have alignment data

presence_absence_df = pd.DataFrame({'loci':x_labels,'presence':0})

for locus in list_of_loci_with_alignments:

row_index = presence_absence_df[presence_absence_df.loci == locus].index

presence_absence_df.loc[row_index,'presence'] = 1

data_2_contig_alignment = np.array(presence_absence_df.presence)

data_2_y_labels = np.array('contig alignment')

#_______________________________3. Reference-assembly Data__________________________

# Get the data as pandas dataframe

unsorted_read_cov_data = pd.read_csv(read_cov_file, sep = '\t',index_col=0)

locus_selection=False

if 'sum_per_locus' in unsorted_read_cov_data.columns:

unsorted_read_cov_data = unsorted_read_cov_data.iloc[:,:-1]

locus_selection=True

# sort columns in df

temp_read_cov_data = unsorted_read_cov_data[sorted(unsorted_read_cov_data.columns)].sort_index()

# add row of 0's for all missing loci

loci_in_df = list(temp_read_cov_data.index)

for locus in list(x_labels):

if locus not in loci_in_df:

temp_read_cov_data.loc[locus] = [0.0]*len(temp_read_cov_data.columns)

# sort by index again

read_cov_data = temp_read_cov_data.sort_index()

# turn df into matrix

data_3_read_cov = np.matrix(read_cov_data).T

# lets use the same labels as for the contig data

data_3_y_labels = np.core.defchararray.replace(data_1_y_labels, 'contigs', 'coverage')

#___________________________Combine all Data___________________________________

combined_data = np.vstack([data_1_contig_present, data_2_contig_alignment,data_3_read_cov])

tmp_combined_y_labels = np.append(data_1_y_labels,data_2_y_labels)

combined_y_labels = np.append(tmp_combined_y_labels,data_3_y_labels)

height,width = combined_data.shape

# Define the range for the heatmap, above the maximum everything will be colored in the highest color

norm=None

if norm_value:

norm = mpl.colors.Normalize(vmin=0, vmax=norm_value)

if locus_selection and reduce:

print('Reducing final matrix to selected loci.')

# Reduce the data matrix to only those columns for which we have extracted values in the selected loci

boolean = combined_data[-1]>0

reduced_data = np.matrix([np.matrix.tolist(ind[boolean])[0] for ind in combined_data])

# also print a text file with the loci indeces and the corresponding loci names

new_locus_list = x_labels[np.array(boolean)[0]]

locus_index_overview = pd.DataFrame({'loci':new_locus_list})

locus_index_overview.to_csv(os.path.join(workdir,'locus_index_overview.txt'),sep='\t',header=False)

# continue with plotting

fig = plt.figure(figsize=(13.5,8))

for i,m in enumerate(reduced_data):

ax = plt.subplot(height, 1, i+1)

ax.tick_params(left='off',bottom='off',labelleft='off')

# Only plot x-axis for last row

if not i == height-1:

ax.xaxis.set_major_formatter(plt.NullFormatter())

#plt.axis("off")

if combined_y_labels[i] == 'contig alignment':

plt.imshow(reduced_data[i], aspect='auto', cmap='Greens_r', origin='lower')

elif 'contigs' in combined_y_labels[i]:

plt.imshow(reduced_data[i], aspect='auto', cmap='GnBu', origin='lower')

else:

plt.imshow(reduced_data[i], aspect='auto', cmap='hot_r',norm=norm, origin='lower')#,clim=(0.0, 10))

pos = list(ax.get_position().bounds)

fig.text(pos[0] - 0.01, pos[1], combined_y_labels[i], horizontalalignment='right')

plt.xlabel('exon index')

else:

# print a text file with the loci indeces and the corresponding loci names

new_locus_list = x_labels

locus_index_overview = pd.DataFrame({'loci':new_locus_list})

locus_index_overview.to_csv(os.path.join(workdir,'locus_index_overview.txt'),sep='\t',header=False)

if not number_of_rows:

#_______________________________Plot Combined Data_____________________________

fig = plt.figure(figsize=(20,8))

#fig.subplots_adjust(top=1, bottom=0.0, left=0.2, right=0.99)

for i,m in enumerate(combined_data):

ax = plt.subplot(height, 1, i+1)

ax.tick_params(left='off',bottom='off',labelleft='off')

# Only plot x-axis for last row

if not i == height-1:

ax.xaxis.set_major_formatter(plt.NullFormatter())

#plt.axis("off")

if combined_y_labels[i] == 'contig alignment':

plt.imshow(combined_data[i], aspect='auto', cmap='Greens', origin='lower')

elif 'contigs' in combined_y_labels[i]:

plt.imshow(combined_data[i], aspect='auto', cmap='GnBu', origin='lower')

else:

plt.imshow(combined_data[i], aspect='auto', cmap='hot_r',norm=norm, origin='lower')#,clim=(0.0, 10))

pos = list(ax.get_position().bounds)

fig.text(pos[0] - 0.01, pos[1], combined_y_labels[i], horizontalalignment='right')

plt.xlabel('exon index')

#plt.colorbar()

elif number_of_rows:

images = []

#_______________________________Plot Split Data_____________________________

# Split dataset for better readability

columns_per_row = int(combined_data.shape[1]/number_of_rows)

remainder = combined_data.shape[1]%number_of_rows

# start and end of first row, adding the remainder to the first row

b = 0

n = columns_per_row+remainder

subset_dict = {}

# iterate through chunks

for i in range(number_of_rows):

data_chunk = combined_data[:,b:n]

subset_dict.setdefault('%i,%i' %(b,n),data_chunk)

b=n

n+=columns_per_row

for j in subset_dict.keys():

split_data = subset_dict[j]

data_range = j.split(',')

num_x_labels = np.arange(int(data_range[0]),int(data_range[-1]))

fig = plt.figure(figsize=(20,8))

#fig.subplots_adjust(top=1, bottom=0.0, left=0.2, right=0.99)

for i,m in enumerate(split_data):

ax = plt.subplot(height, 1, i+1)

ax.tick_params(left='off',bottom='off',labelleft='off')

# Only plot x-axis for last row

if not i == height-1:

ax.xaxis.set_major_formatter(plt.NullFormatter())

#plt.axis("off")

if combined_y_labels[i] == 'contig alignment':

plt.imshow(split_data[i], aspect='auto', cmap='Greens', origin='lower')

elif 'contigs' in combined_y_labels[i]:

plt.imshow(split_data[i], aspect='auto', cmap='GnBu', origin='lower')

else:

plt.imshow(split_data[i], aspect='auto', cmap='hot_r',norm=norm, origin='lower')#,clim=(0.0, 10))

pos = list(ax.get_position().bounds)

fig.text(pos[0] - 0.01, pos[1], combined_y_labels[i], horizontalalignment='right')

# make sure to have 10 ticks on the x-axis (ensured by dividing the total lenght by 9 and using the resulting value as stepsize)

tick_step_size = split_data[i].shape[1]/9

# get the desired indeces of the x-values that shall carry ticks on x-axis

xi = np.arange(0, split_data[i].shape[1],int(tick_step_size))

# get the corresponding x-values from the num_x_labels variable (dicitonary keys)

x = np.arange(num_x_labels[0], num_x_labels[-1], int(tick_step_size))

plt.xticks(xi,x)

plt.xlabel('exon index')

fig.savefig(os.path.join(workdir,'contig_exon_coverage_matrix_%s.png'%j), dpi = 500)

images.append(fig)

if not number_of_rows:

return fig

else:

#________________________________Plot Legend___________________________________

# Make a figure and axes with dimensions as desired.

mpl.rcParams.update({'font.size': font_size})

fig = plt.figure(figsize=(width, height))

# the values stand for [x0,x1,width,height] --> all in relation to total size as given by 'figsize='

ax1 = fig.add_axes([0.1,0.05,.3,.9])

# Set the colormap and norm to correspond to the data for which

# the colorbar will be used.

cmap = mpl.cm.hot_r

norm = mpl.colors.Normalize(vmin=min_value, vmax=max_value)

# plot a basic continuous colorbar with ticks and labels.

cb1 = mpl.colorbar.ColorbarBase(ax1, cmap=cmap,

norm=norm,

orientation='vertical')

cb1.set_label('Read coverage')

cb1.ax.set_yticklabels(['0','2','4','6','8','>10'])

#plt.show()

mpl.rcParams.update({'font.size': font_size})

fig = plt.figure(figsize=(width, height))

# the values stand for [x0,x1,width,height] --> all in relation to total size as given by 'figsize='

ax1 = fig.add_axes([0.1,0.1,.4,.8])

# Set the colormap and norm to correspond to the data for which

cmap = mpl.cm.binary

cmaplist = [cmap(i) for i in range(cmap.N)]

# define black and white manually, make beginning to center black and center to end white

center = int(len(cmaplist)/2)

cmaplist[0:center + 1] = [(1.0,1.0,1.0,1.0)] * ((center + 1) - 0)

cmaplist[center+1:] = [(0.0,0.0,0.0,1.0)]*(len(cmaplist)+1-center)

# create the new map

cmap = cmap.from_list('Custom cmap', cmaplist)

bounds = np.linspace(0,1,3)

norm = mpl.colors.BoundaryNorm(bounds, cmap.N)

# plot a basic continuous colorbar with ticks and labels.

cb1 = mpl.colorbar.ColorbarBase(ax1, cmap=cmap,

norm=norm,

orientation='vertical',spacing='proportional', ticks=[0.25,0.75], boundaries=bounds)

cb1.ax.set_yticklabels(['Absence','Presence'])

#cb1.set_label('Presence/Absence')

if tick_labels == 'auto':

labels = np.arange(0,n_categories)

else:

labels=tick_labels

mpl.rcParams.update({'font.size': font_size})

fig = plt.figure(figsize=(width, height))

# the values stand for [x0,x1,width,height] --> all in relation to total size as given by 'figsize='

ax = fig.add_axes([x0,x1,plot_width,plot_height])

# define a colormap by providing the beginning and end color and number of categories between these colors

cmap = mpl.colors.LinearSegmentedColormap.from_list("my_colormap", (color1, color2), N=n_categories, gamma=1.0)

bounds = np.linspace(0,1,n_categories+1)

# the ticks should be in the cneter between any two bounds, so move all 1/2 unit up

ticks = bounds[:-1]+(1/n_categories/2)

norm = mpl.colors.Normalize(0, norm_max)

if colormap:

cb = mpl.colorbar.ColorbarBase(ax, cmap=colormap,norm=norm,

orientation='vertical',spacing='proportional', ticks=ticks, boundaries=bounds)

else:

cb = mpl.colorbar.ColorbarBase(ax, cmap=cmap,norm=norm,

orientation='vertical',spacing='proportional', ticks=ticks, boundaries=bounds)

cb.ax.set_yticklabels(labels)

if plot_label:

cb.set_label(plot_label)

number_of_rows = int(combined_data.shape[1]/columns_per_row)

b = 0

n = columns_per_row

subset_dict = {}

# iterate through chunks

for i in range(number_of_rows):

data_chunk = combined_data[:,b:n]

subset_dict.setdefault('%i,%i' %(b,n),data_chunk)

b=n

n+=columns_per_row

# take care of the remainder, if there is some left

if b<combined_data.shape[1]:

n=combined_data.shape[1]

data_chunk = combined_data[:,b:n]

import glob

import os

import numpy as np

import scipy.misc

import matplotlib.pyplot as plt

#%matplotlib inline

# get all subplot files:

work_dir = "../../data/processed/remapped_reads/"

images = []

for file in glob.glob(os.path.join(work_dir,"*.png")):

images.append(file)

# convert images into data objects

img_data = []

for img in images:

img_data.append(scipy.misc.imread(img))

# plot all images in one window

plt.figure(figsize=(200,100))

columns = 1

for i, image in enumerate(np.array(img_data)):

plt.subplot(len(images) / columns + 1, columns, i + 1)

plt.axis('off')

plt.imshow(image)