Skip to content

Fixed issues with python3 and matplotlib #14

New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

Sign up for GitHub

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

Open
milkschen wants to merge 2 commits into
base: master
Choose a base branch
from
Open

Fixed issues with python3 and matplotlib #14

Changes from all commits
Commits
Show all changes
2 commits
Select commit Hold shift + click to select a range
4ee18a7
fixed issues with python3 and latest version of matplotlib; this work…
milkschen Mar 26, 2020
2daf43d
uses flexible ylim (lower part); added --label feature so that the la…
milkschen Mar 26, 2020
File filter

Filter by extension

Filter by extension
Conversations
Failed to load comments.
Loading
Jump to
Jump to file
Failed to load files.
Loading
Diff view
Diff view
45 changes: 33 additions & 12 deletions varmatch
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -21,7 +21,8 @@ limitations under the License.
Chen Sun(chensun@cse.psu.edu)
Paul Medvedev(pashadag@cse.psu.edu)
"""
from __future__ import print_function
# from __future__ import print_function
exec('from __future__ import print_function')
import sys
versionError = "You are using an old version of python, please upgrade to python 2.7+\n"
if sys.hexversion < 0x02070000:
Expand Down Expand Up @@ -56,6 +57,7 @@ citation = 'Please cite our paper.'
parser = argparse.ArgumentParser(prog="varmatch", epilog = citation, formatter_class=lambda prog: SmartFormatter(prog,max_help_position=8))
parser.add_argument('-b', '--baseline', required=True, metavar='File', help = 'baseline variant VCF filename')
parser.add_argument('-q', '--query', nargs='+', metavar='File List', help = 'query variant VCF filename')
parser.add_argument('-l', '--label', nargs='+', metavar='Label List', help = 'query variant VCF label showing in figures', default=None)
parser.add_argument('-g', '--genome', required=True, metavar='File', help= 'genome sequence FASTA filename')
parser.add_argument('-o', '--output', help='output directory', metavar='DIRECTORY',default='./output')

Expand Down Expand Up @@ -325,23 +327,26 @@ def parse_stat(output_prefix):
return baseline_num, query_num, x, y, z, no_filter_table, sensitivity_table, specificity_table


def create_table_by_matchingid_from_by_query(table_list, matching_list, query_number):
def create_table_by_matchingid_from_by_query(table_list, matching_list, query_number, query_label_list):
table_by_matchingid = []
for matching_index in range(len(matching_list)):
matching_table = []
title = ['Query Id', 'Baseline Match Number', 'Query Match Number', 'Recall(%)', 'Precision(%)']
matching_table.append(title)
for table_index in range(len(table_list)):
raw_row = table_list[table_index][matching_index]
new_row = ['Query' + str(table_index+1)]
if query_label_list:
new_row = [query_label_list[table_index]]
else:
new_row = ['Query' + str(table_index+1)]
new_row += raw_row[4:]
matching_table.append(new_row)
table_by_matchingid.append(matching_table)
return table_by_matchingid


# all html and picture are created from stat file, not parameters
def create_stat_html(query_list, output_prefix_list):
def create_stat_html(query_list, output_prefix_list, query_label_list):
global output_dir
html_filename = output_dir + '/stat.html'
html_file = open(html_filename, 'w')
Expand Down Expand Up @@ -384,15 +389,21 @@ def create_stat_html(query_list, output_prefix_list):
import matplotlib.pyplot as plt
axes = plt.gca()
#axes.set_xlim([xmin,xmax])
axes.set_ylim([0,100])
# axes.set_ylim([0,100])
min_sensitivity = 100
for i in range(exp_num):
marker_id = i % len(marker_list)
marker_sign = marker_list[marker_id]
label_sign = 'Query ' + str(i+1)
if query_label_list:
label_sign = query_label_list[i]
else:
label_sign = 'Query ' + str(i+1)
x = np.array(range(len(label_list[0])))
plt.xticks(x, label_list[0])
plt.plot(x, sensitivity_list[i], marker = marker_sign, linestyle = '-', label = label_sign)
min_sensitivity = min(min_sensitivity, min(sensitivity_list[i]))

axes.set_ylim([min_sensitivity * 0.9,100])
plt.xlabel('Matching Id')
plt.ylabel('Recall(%)')
#plt.title('Sensitivity of Queries under Different Matching Parameters')
Expand All @@ -403,14 +414,20 @@ def create_stat_html(query_list, output_prefix_list):
axes = plt.gca()
#axes.set_xlim([xmin,xmax])
axes.set_ylim([0,100])
min_precision = 100
for i in range(exp_num):
marker_id = i % len(marker_list)
marker_sign = marker_list[marker_id]
label_sign = 'Query ' + str(i+1)
if query_label_list:
label_sign = query_label_list[i]
else:
label_sign = 'Query ' + str(i+1)
x = np.array(range(len(label_list[0])))
plt.xticks(x, label_list[0])
plt.plot(x, specificity_list[i], marker = marker_sign, linestyle = '-', label = label_sign)
min_precision = min(min_precision, min(specificity_list[i]))

axes.set_ylim([min_precision * 0.9,100])
plt.xlabel('Matching Id')
plt.ylabel('Precision(%)')
#plt.title('Specificity of Queries under Different Matching Parameters')
Expand Down Expand Up @@ -448,7 +465,7 @@ def create_stat_html(query_list, output_prefix_list):
# sensitivity and specificity analysis by matching id
html_file.write('<h2><Sensitivity and Specificity Analysis by Matching Id/h2>')

table_by_matchingid = create_table_by_matchingid_from_by_query(table_list, label_list[0], exp_num)
table_by_matchingid = create_table_by_matchingid_from_by_query(table_list, label_list[0], exp_num, query_label_list)
for i in range(len(label_list[0])):
html_file.write('<div class="box">')
html_file.write('<h3>Matching Id: ' + label_list[0][i] + '</h3>'+'\n')
Expand Down Expand Up @@ -494,7 +511,10 @@ def create_stat_html(query_list, output_prefix_list):
x.sort()
y.sort(reverse=True)
y = y[::-1]
label_sign = 'Query ' + str(j+1)
if query_label_list:
label_sign = query_label_list[j]
else:
label_sign = 'Query ' + str(j+1)
plt.plot(x,y, label = label_sign)

#x = [0.0, 1.0]
Expand All @@ -518,7 +538,7 @@ def create_stat_html(query_list, output_prefix_list):
plt.clf()

colormap = plt.cm.gist_ncar
plt.gca().set_color_cycle([colormap(k) for k in np.linspace(0, 0.9, len(parameter_table))])
plt.gca().set_prop_cycle(color = [colormap(k) for k in np.linspace(0, 0.9, len(parameter_table))])

for j in range(len(parameter_table)-1):
x = sensitivity_table_list[i][j]
Expand Down Expand Up @@ -583,8 +603,9 @@ def main():
temp_dir = output_dir + '/temp'

query_list = args.query
label_list = args.label

multiple_compare(args.baseline, query_list, args.genome)
# multiple_compare(args.baseline, query_list, args.genome)

if args.score_scheme == '3':
exit()
Expand All @@ -593,7 +614,7 @@ def main():
for i in range(len(query_list)):
output_prefix_list.append('query'+str(i+1))

create_stat_html(query_list, output_prefix_list)
create_stat_html(query_list, output_prefix_list, label_list)

if __name__ == '__main__':
main()