def main():
option_parser, opts, args = parse_command_line_parameters(**script_info)
# Create the output dir if it doesn't already exist.
out_dir = opts.output_dir
try:
create_dir(out_dir)
except:
option_parser.error("Could not create or access output directory "
"specified with the -o option.")
map_f = open(opts.mapping_fp, 'U')
dm_f = open(opts.distance_matrix_fp, 'U')
fields = map(strip, opts.fields.split(','))
fields = [field.strip('"').strip("'") for field in fields]
color_individual_within_by_field = opts.color_individual_within_by_field
results = make_distance_boxplots(dm_f, map_f, fields, width=opts.width,
height=opts.height, suppress_all_within=opts.suppress_all_within,
suppress_all_between=opts.suppress_all_between,
suppress_individual_within=opts.suppress_individual_within,
suppress_individual_between=opts.suppress_individual_between,
y_min=opts.y_min, y_max=opts.y_max,
whisker_length=opts.whisker_length, box_width=opts.box_width,
box_color=opts.box_color,
color_individual_within_by_field=color_individual_within_by_field,
sort=opts.sort)
for field, plot_figure, plot_data, plot_labels, plot_colors in results:
output_plot_fp = join(out_dir, "%s_Distances.%s" %
(field, opts.imagetype))
plot_figure.savefig(output_plot_fp, format=opts.imagetype,
transparent=opts.transparent)
if not opts.suppress_significance_tests:
sig_tests_f = open(join(out_dir, "%s_Stats.txt" % field), 'w')
sig_tests_results = all_pairs_t_test(plot_labels, plot_data,
tail_type=opts.tail_type,
num_permutations=opts.num_permutations)
sig_tests_f.write(sig_tests_results)
sig_tests_f.close()
if opts.save_raw_data:
# Write the raw plot data into a tab-delimited file.
assert(len(plot_labels) == len(plot_data))
raw_data_fp = join(out_dir, "%s_Distances.txt" % field)
raw_data_f = open(raw_data_fp, 'w')
for label, data in zip(plot_labels, plot_data):
raw_data_f.write(label.replace(" ", "_") + "\t")
raw_data_f.write("\t".join(map(str, data)))
raw_data_f.write("\n")
raw_data_f.close()
def main():
option_parser, opts, args = parse_command_line_parameters(**script_info)
# Create the output dir if it doesn't already exist.
try:
create_dir(opts.output_dir)
except:
option_parser.error("Could not create or access output directory " "specified with the -o option.")
# Parse the distance matrix and mapping file.
try:
dist_matrix_header, dist_matrix = parse_distmat(open(opts.distance_matrix_fp, "U"))
except:
option_parser.error(
"This does not look like a valid distance matrix "
"file. Please supply a valid distance matrix file using the -d "
"option."
)
try:
mapping, mapping_header, mapping_comments = parse_mapping_file(open(opts.mapping_fp, "U"))
except QiimeParseError:
option_parser.error(
"This does not look like a valid metadata mapping "
"file. Please supply a valid mapping file using the -m option."
)
# Make sure the y_min and y_max options make sense, as they can be either
# 'auto' or a number.
y_min = opts.y_min
y_max = opts.y_max
try:
y_min = float(y_min)
except ValueError:
if y_min == "auto":
y_min = None
else:
option_parser.error("The --y_min option must be either a number " "or 'auto'.")
try:
y_max = float(y_max)
except ValueError:
if y_max == "auto":
y_max = None
else:
option_parser.error("The --y_max option must be either a number " "or 'auto'.")
# Parse the field states that will be compared to every other field state.
comparison_field_states = opts.comparison_groups
comparison_field_states = map(strip, comparison_field_states.split(","))
comparison_field_states = [field_state.strip('"').strip("'") for field_state in comparison_field_states]
if comparison_field_states is None:
option_parser.error("You must provide at least one field state to " "compare (using the -c option).")
# Get distance comparisons between each field state and each of the
# comparison field states.
field = opts.field
comparison_groupings = get_field_state_comparisons(
dist_matrix_header, dist_matrix, mapping_header, mapping, field, comparison_field_states
)
# Grab a list of all field states that had the comparison field states
# compared against them. These will be plotted along the x-axis.
field_states = comparison_groupings.keys()
def custom_comparator(x, y):
try:
num_x = float(x)
num_y = float(y)
return int(num_x - num_y)
except:
if x < y:
return -1
elif x > y:
return 1
else:
return 0
# Sort the field states as numbers if the elements are numbers, else sort
# them lexically.
field_states.sort(custom_comparator)
# If the label type is numeric, get a list of all field states in sorted
# numeric order. These will be used to determine the spacing of the
# field state 'points' along the x-axis.
x_spacing = None
if opts.label_type == "numeric":
try:
x_spacing = sorted(map(float, field_states))
except:
option_parser.error(
"The 'numeric' label type is invalid because "
"not all field states could be converted into "
"numbers. Please specify a different label "
"type."
)
# Accumulate the data for each field state 'point' along the x-axis.
plot_data = []
plot_x_axis_labels = []
for field_state in field_states:
field_state_data = []
for comp_field_state in comparison_field_states:
field_state_data.append(comparison_groupings[field_state][comp_field_state])
plot_data.append(field_state_data)
plot_x_axis_labels.append(field_state)
# Plot the data and labels.
plot_title = "Distance Comparisons"
plot_x_label = field
plot_y_label = "Distance"
# If we are creating a bar chart or box plot, grab a list of good data
# colors to use.
plot_type = opts.plot_type
plot_colors = None
if plot_type == "bar" or plot_type == "box":
plot_colors = [matplotlib_rgb_color(data_colors[color].toRGB()) for color in data_color_order]
assert plot_data, "Error: there is no data to plot!"
width = opts.width
height = opts.height
if width <= 0 or height <= 0:
option_parser.error("The specified width and height of the image must " "be greater than zero.")
plot_figure = grouped_distributions(
opts.plot_type,
plot_data,
x_values=x_spacing,
data_point_labels=plot_x_axis_labels,
distribution_labels=comparison_field_states,
distribution_markers=plot_colors,
x_label=plot_x_label,
y_label=plot_y_label,
title=plot_title,
x_tick_labels_orientation=opts.x_tick_labels_orientation,
y_min=y_min,
y_max=y_max,
whisker_length=opts.whisker_length,
error_bar_type=opts.error_bar_type,
distribution_width=opts.distribution_width,
figure_width=width,
figure_height=height,
)
# Save the plot in the specified format.
output_plot_fp = join(opts.output_dir, "%s_Distance_Comparisons.%s" % (field, opts.imagetype))
plot_figure.savefig(output_plot_fp, format=opts.imagetype, transparent=opts.transparent)
if not opts.suppress_significance_tests:
sig_tests_f = open(join(opts.output_dir, "%s_Stats.txt" % field), "w")
# Rearrange the plot data into a format suitable for all_pairs_t_test.
sig_tests_labels = []
sig_tests_data = []
for data_point, data_point_label in zip(plot_data, plot_x_axis_labels):
for dist, comp_field in zip(data_point, comparison_field_states):
sig_tests_labels.append("%s vs %s" % (data_point_label, comp_field))
sig_tests_data.append(dist)
sig_tests_results = all_pairs_t_test(
sig_tests_labels, sig_tests_data, tail_type=opts.tail_type, num_permutations=opts.num_permutations
)
sig_tests_f.write(sig_tests_results)
sig_tests_f.close()
if opts.save_raw_data:
# Write the raw plot data into a tab-delimited file, where each line
# has the distances between a comparison group and another field state
# 'point' along the x-axis.
assert len(plot_x_axis_labels) == len(plot_data), (
"The number of " + "labels do not match the number of points along the x-axis."
)
raw_data_fp = join(opts.output_dir, "%s_Distance_Comparisons.txt" % field)
raw_data_f = open(raw_data_fp, "w")
raw_data_f.write("#ComparisonGroup\tFieldState\tDistances\n")
for label, data in zip(plot_x_axis_labels, plot_data):
assert len(comparison_field_states) == len(data), (
"The "
+ "number of specified comparison groups does not match "
+ "the number of groups found at the current point along "
+ "the x-axis."
)
for comp_field_state, comp_grp_data in zip(comparison_field_states, data):
raw_data_f.write(comp_field_state + "\t" + label + "\t" + "\t".join(map(str, comp_grp_data)) + "\n")
raw_data_f.close()
def main():
option_parser, opts, args = parse_command_line_parameters(**script_info)
# Create the output dir if it doesn't already exist.
out_dir = opts.output_dir
try:
create_dir(out_dir)
except:
option_parser.error("Could not create or access output directory "
"specified with the -o option.")
map_f = open(opts.mapping_fp, 'U')
dm_f = open(opts.distance_matrix_fp, 'U')
fields = map(strip, opts.fields.split(','))
fields = [field.strip('"').strip("'") for field in fields]
color_individual_within_by_field = opts.color_individual_within_by_field
results = make_distance_boxplots(
dm_f,
map_f,
fields,
width=opts.width,
height=opts.height,
suppress_all_within=opts.suppress_all_within,
suppress_all_between=opts.suppress_all_between,
suppress_individual_within=opts.suppress_individual_within,
suppress_individual_between=opts.suppress_individual_between,
y_min=opts.y_min,
y_max=opts.y_max,
whisker_length=opts.whisker_length,
box_width=opts.box_width,
box_color=opts.box_color,
color_individual_within_by_field=color_individual_within_by_field,
sort=opts.sort)
for field, plot_figure, plot_data, plot_labels, plot_colors in results:
output_plot_fp = join(out_dir,
"%s_Distances.%s" % (field, opts.imagetype))
plot_figure.savefig(output_plot_fp,
format=opts.imagetype,
transparent=opts.transparent)
if not opts.suppress_significance_tests:
sig_tests_f = open(join(out_dir, "%s_Stats.txt" % field), 'w')
sig_tests_results = all_pairs_t_test(
plot_labels,
plot_data,
tail_type=opts.tail_type,
num_permutations=opts.num_permutations)
sig_tests_f.write(sig_tests_results)
sig_tests_f.close()
if opts.save_raw_data:
# Write the raw plot data into a tab-delimited file.
assert (len(plot_labels) == len(plot_data))
raw_data_fp = join(out_dir, "%s_Distances.txt" % field)
raw_data_f = open(raw_data_fp, 'w')
for label, data in zip(plot_labels, plot_data):
raw_data_f.write(label.replace(" ", "_") + "\t")
raw_data_f.write("\t".join(map(str, data)))
raw_data_f.write("\n")
raw_data_f.close()
def main():
option_parser, opts, args = parse_command_line_parameters(**script_info)
# Create the output dir if it doesn't already exist.
try:
create_dir(opts.output_dir)
except:
option_parser.error("Could not create or access output directory "
"specified with the -o option.")
# Parse the distance matrix and mapping file.
try:
dist_matrix_header, dist_matrix = parse_distmat(
open(opts.distance_matrix_fp, 'U'))
except:
option_parser.error(
"This does not look like a valid distance matrix "
"file. Please supply a valid distance matrix file using the -d "
"option.")
try:
mapping, mapping_header, mapping_comments = parse_mapping_file(
open(opts.mapping_fp, 'U'))
except QiimeParseError:
option_parser.error(
"This does not look like a valid metadata mapping "
"file. Please supply a valid mapping file using the -m option.")
# Make sure the y_min and y_max options make sense, as they can be either
# 'auto' or a number.
y_min = opts.y_min
y_max = opts.y_max
try:
y_min = float(y_min)
except ValueError:
if y_min == 'auto':
y_min = None
else:
option_parser.error("The --y_min option must be either a number "
"or 'auto'.")
try:
y_max = float(y_max)
except ValueError:
if y_max == 'auto':
y_max = None
else:
option_parser.error("The --y_max option must be either a number "
"or 'auto'.")
# Parse the field states that will be compared to every other field state.
comparison_field_states = opts.comparison_groups
comparison_field_states = map(strip, comparison_field_states.split(','))
comparison_field_states = [
field_state.strip('"').strip("'")
for field_state in comparison_field_states
]
if comparison_field_states is None:
option_parser.error("You must provide at least one field state to "
"compare (using the -c option).")
# Get distance comparisons between each field state and each of the
# comparison field states.
field = opts.field
comparison_groupings = get_field_state_comparisons(
dist_matrix_header, dist_matrix, mapping_header, mapping, field,
comparison_field_states)
# Grab a list of all field states that had the comparison field states
# compared against them. These will be plotted along the x-axis.
field_states = comparison_groupings.keys()
def custom_comparator(x, y):
try:
num_x = float(x)
num_y = float(y)
return int(num_x - num_y)
except:
if x < y:
return -1
elif x > y:
return 1
else:
return 0
# Sort the field states as numbers if the elements are numbers, else sort
# them lexically.
field_states.sort(custom_comparator)
# If the label type is numeric, get a list of all field states in sorted
# numeric order. These will be used to determine the spacing of the
# field state 'points' along the x-axis.
x_spacing = None
if opts.label_type == "numeric":
try:
x_spacing = sorted(map(float, field_states))
except:
option_parser.error("The 'numeric' label type is invalid because "
"not all field states could be converted into "
"numbers. Please specify a different label "
"type.")
# Accumulate the data for each field state 'point' along the x-axis.
plot_data = []
plot_x_axis_labels = []
for field_state in field_states:
field_state_data = []
for comp_field_state in comparison_field_states:
field_state_data.append(
comparison_groupings[field_state][comp_field_state])
plot_data.append(field_state_data)
plot_x_axis_labels.append(field_state)
# Plot the data and labels.
plot_title = "Distance Comparisons"
plot_x_label = field
plot_y_label = "Distance"
# If we are creating a bar chart or box plot, grab a list of good data
# colors to use.
plot_type = opts.plot_type
plot_colors = None
if plot_type == "bar" or plot_type == "box":
plot_colors = [
matplotlib_rgb_color(data_colors[color].toRGB())
for color in data_color_order
]
assert plot_data, "Error: there is no data to plot!"
width = opts.width
height = opts.height
if width <= 0 or height <= 0:
option_parser.error("The specified width and height of the image must "
"be greater than zero.")
plot_figure = generate_comparative_plots(
opts.plot_type,
plot_data,
x_values=x_spacing,
data_point_labels=plot_x_axis_labels,
distribution_labels=comparison_field_states,
distribution_markers=plot_colors,
x_label=plot_x_label,
y_label=plot_y_label,
title=plot_title,
x_tick_labels_orientation=opts.x_tick_labels_orientation,
y_min=y_min,
y_max=y_max,
whisker_length=opts.whisker_length,
error_bar_type=opts.error_bar_type,
distribution_width=opts.distribution_width,
figure_width=width,
figure_height=height)
# Save the plot in the specified format.
output_plot_fp = join(
opts.output_dir,
"%s_Distance_Comparisons.%s" % (field, opts.imagetype))
plot_figure.savefig(output_plot_fp,
format=opts.imagetype,
transparent=opts.transparent)
if not opts.suppress_significance_tests:
sig_tests_f = open(join(opts.output_dir, "%s_Stats.txt" % field), 'w')
# Rearrange the plot data into a format suitable for all_pairs_t_test.
sig_tests_labels = []
sig_tests_data = []
for data_point, data_point_label in zip(plot_data, plot_x_axis_labels):
for dist, comp_field in zip(data_point, comparison_field_states):
sig_tests_labels.append('%s vs %s' %
(data_point_label, comp_field))
sig_tests_data.append(dist)
sig_tests_results = all_pairs_t_test(
sig_tests_labels,
sig_tests_data,
tail_type=opts.tail_type,
num_permutations=opts.num_permutations)
sig_tests_f.write(sig_tests_results)
sig_tests_f.close()
if opts.save_raw_data:
# Write the raw plot data into a tab-delimited file, where each line
# has the distances between a comparison group and another field state
# 'point' along the x-axis.
assert (len(plot_x_axis_labels) == len(plot_data)), "The number of " +\
"labels do not match the number of points along the x-axis."
raw_data_fp = join(opts.output_dir,
"%s_Distance_Comparisons.txt" % field)
raw_data_f = open(raw_data_fp, 'w')
raw_data_f.write("#ComparisonGroup\tFieldState\tDistances\n")
for label, data in zip(plot_x_axis_labels, plot_data):
assert (len(comparison_field_states) == len(data)), "The " +\
"number of specified comparison groups does not match " +\
"the number of groups found at the current point along " +\
"the x-axis."
for comp_field_state, comp_grp_data in zip(comparison_field_states,
data):
raw_data_f.write(comp_field_state + "\t" + label + "\t" +
"\t".join(map(str, comp_grp_data)) + "\n")
raw_data_f.close()
def main():
option_parser, opts, args = parse_command_line_parameters(**script_info)
# Create the output dir if it doesn't already exist.
try:
create_dir(opts.output_dir)
except:
option_parser.error("Could not create or access output directory "
"specified with the -o option.")
# Parse the distance matrix and mapping file.
try:
dist_matrix_header, dist_matrix = parse_distmat(
open(opts.distance_matrix_fp, 'U'))
except:
option_parser.error("This does not look like a valid distance matrix "
"file. Please supply a valid distance matrix file using the -d "
"option.")
try:
mapping, mapping_header, mapping_comments = parse_mapping_file(
open(opts.mapping_fp, 'U'))
except QiimeParseError:
option_parser.error("This does not look like a valid metadata mapping "
"file. Please supply a valid mapping file using the -m option.")
fields = opts.fields
fields = map(strip, fields.split(','))
fields = [field.strip('"').strip("'") for field in fields]
if fields is None:
option_parser.error("You must provide at least one field using the -f "
"option.")
# Make sure each field is in the mapping file.
for field in fields:
if field not in mapping_header:
option_parser.error("The field '%s' is not in the provided "
"mapping file. Please supply correct fields (using the -f "
"option) corresponding to fields in the mapping file."
% field)
# Make sure the y_min and y_max options make sense, as they can be either
# 'auto' or a number.
y_min = opts.y_min
y_max = opts.y_max
try:
y_min = float(y_min)
except ValueError:
if y_min == 'auto':
y_min = None
else:
option_parser.error("The --y_min option must be either a number "
"or 'auto'.")
try:
y_max = float(y_max)
except ValueError:
if y_max == 'auto':
y_max = None
else:
option_parser.error("The --y_max option must be either a number "
"or 'auto'.")
# Generate the various boxplots, depending on what the user wanted
# suppressed. Add them all to one encompassing plot.
for field in fields:
plot_data = []
plot_labels = []
if not opts.suppress_all_within:
plot_data.append(get_all_grouped_distances(dist_matrix_header,
dist_matrix, mapping_header, mapping, field, within=True))
plot_labels.append("All within %s" % field)
if not opts.suppress_all_between:
plot_data.append(get_all_grouped_distances(dist_matrix_header,
dist_matrix, mapping_header, mapping, field, within=False))
plot_labels.append("All between %s" % field)
if not opts.suppress_individual_within:
within_dists = get_grouped_distances(dist_matrix_header,
dist_matrix, mapping_header, mapping, field, within=True)
for grouping in within_dists:
plot_data.append(grouping[2])
plot_labels.append("%s vs. %s" % (grouping[0], grouping[1]))
if not opts.suppress_individual_between:
between_dists = get_grouped_distances(dist_matrix_header,
dist_matrix, mapping_header, mapping, field, within=False)
for grouping in between_dists:
plot_data.append(grouping[2])
plot_labels.append("%s vs. %s" % (grouping[0], grouping[1]))
# We now have our data and labels ready, so plot them!
assert (len(plot_data) == len(plot_labels)), "The number " +\
"of boxplot labels does not match the number of " +\
"boxplots."
if plot_data:
if opts.sort:
# Sort our plot data in order of increasing median.
sorted_data = []
for label, distribution in zip(plot_labels, plot_data):
sorted_data.append((label, distribution,
median(distribution)))
sorted_data.sort(key=itemgetter(2))
plot_labels = []
plot_data = []
for label, distribution, median_value in sorted_data:
plot_labels.append(label)
plot_data.append(distribution)
width = opts.width
height = opts.height
if width is None:
width = len(plot_data) * opts.box_width + 2
if width <= 0 or height <= 0:
option_parser.error("The specified width and height of the "
"image must be greater than zero.")
plot_figure = generate_box_plots(plot_data,
x_tick_labels=plot_labels, title="%s Distances" % field,
x_label="Grouping", y_label="Distance",
x_tick_labels_orientation='vertical', y_min=y_min,
y_max=y_max, whisker_length=opts.whisker_length,
box_width=opts.box_width, box_color=opts.box_color,
figure_width=width, figure_height=height)
output_plot_fp = join(opts.output_dir, "%s_Distances.%s"
% (field, opts.imagetype))
plot_figure.savefig(output_plot_fp, format=opts.imagetype,
transparent=opts.transparent)
else:
option_parser.error("You have chosen to suppress all plots. At "
"least one type of plot must be unsuppressed.")
if not opts.suppress_significance_tests:
sig_tests_f = open(join(
opts.output_dir, "%s_Stats.xls" % field), 'w')
sig_tests_results = all_pairs_t_test(plot_labels, plot_data,
tail_type=opts.tail_type,
num_permutations=opts.num_permutations)
sig_tests_f.write(sig_tests_results)
sig_tests_f.close()
if opts.save_raw_data:
# Write the raw plot data into a tab-delimited file.
assert(len(plot_labels) == len(plot_data))
raw_data_fp = join(opts.output_dir, "%s_Distances.xls"
% field)
raw_data_f = open(raw_data_fp, 'w')
for label, data in zip(plot_labels, plot_data):
raw_data_f.write(label.replace(" ", "_") + "\t")
raw_data_f.write("\t".join(map(str, data)))
raw_data_f.write("\n")
raw_data_f.close()