def main():
option_parser, opts, args =\
parse_command_line_parameters(**script_info)
map_data, map_header, map_comments = parse_mapping_file(open(
opts.map, 'U'))
map_dict = mapping_file_to_dict(map_data, map_header)
distdict = parse_distmat_to_dict(open(opts.distance_matrix, 'U'))
if opts.colorby == None:
colorby_cats = [None]
else:
colorby_idx = map_header.index(opts.colorby)
colorby_cats = list(set([map_data[i][colorby_idx] for\
i in range(len(map_data))]))
textfilename = os.path.splitext(opts.output_path)[0] + '.txt'
text_fh = open(textfilename, 'w')
text_fh.write(opts.axis_category + '\tdistance\tSampleID' + '\n')
colorby_cats.sort()
plt.figure()
for cat_num, cat in enumerate(colorby_cats):
# collect the primary and secondary samples within this category
state1_samids, state2_samids = get_sam_ids(map_data, map_header,
opts.colorby, cat,
opts.primary_state,
opts.secondary_state)
state1_samids =\
list(set(state1_samids).intersection(set(distdict.keys())))
state2_samids =\
list(set(state2_samids).intersection(set(distdict.keys())))
if state1_samids == [] or state2_samids == [] or \
(len(state1_samids) == 1 and state1_samids == state2_samids):
raise RuntimeError("one category of samples didn't have any valid"+\
" distances. try eliminating samples from -p or -s, or changing"+\
" your mapping file with filter_samples_from_otu_table.py")
# go through dmtx
state1_avg_dists = get_avg_dists(state1_samids, state2_samids,
distdict)
# plot
xvals = [float(map_dict[sam][opts.axis_category]) for\
sam in state1_samids]
try:
color = plt.cm.jet(cat_num / (len(colorby_cats) - 1))
except ZeroDivisionError: # only one cat
color = 'b'
plt.scatter(xvals,
state1_avg_dists,
edgecolors=color,
alpha=.5,
facecolors='none')
plt.xlabel(opts.axis_category)
plt.ylabel('average distance')
lines = [str(xvals[i])+'\t'+str(state1_avg_dists[i])+\
'\t'+state1_samids[i]+'\n' for i in range(len(xvals))]
text_fh.writelines(lines)
if opts.colorby != None: plt.legend(colorby_cats)
plt.savefig(opts.output_path)
def test_get_sam_ids(self):
"""set of sample ids in get_sam_ids should be correct"""
map_file = StringIO.StringIO(
"""#SampleID Country AgeYears Family AgeCat
h208A.1 Malawi 0.032854209 h208 Child
h301A.1 Malawi 0.05 h301 Child
h301B.1 Malawi 0.05 h301 Child
USinfTw20.1 USA 0.083333333 USinfTw20 Child
USinfTw20.2 USA 0.083333333 USinfTw20 Child
USinfTw1.1 USA 0.083333333 USinfTw1 Child
h10M Malawi 26 h10 Adult
h68M Malawi 26 h68 Adult
TS25 USA 26 USts9 Adult
TS26 USA 26 USts9 Adult""")
map_data, map_header, comments = parse_mapping_file(map_file)
colorby = 'Country'
cat = 'USA'
primary_state = 'AgeCat:Child'
ids1, ids2 = get_sam_ids(map_data,
map_header,
colorby,
cat,
primary_state,
secondary_state=None)
self.assertEqual(set(ids1),
set(['USinfTw20.1', 'USinfTw20.2', 'USinfTw1.1']))
self.assertEqual(set(ids2), set(['TS25', 'TS26']))
def main():
option_parser, opts, args =\
parse_command_line_parameters(**script_info)
map_data, map_header, map_comments = parse_mapping_file(
open(opts.map, 'U'))
map_dict = mapping_file_to_dict(map_data, map_header)
distdict = parse_distmat_to_dict(open(opts.distance_matrix, 'U'))
if opts.colorby is None:
colorby_cats = [None]
else:
colorby_idx = map_header.index(opts.colorby)
colorby_cats = list(set([map_data[i][colorby_idx] for
i in range(len(map_data))]))
textfilename = os.path.splitext(opts.output_path)[0] + '.txt'
text_fh = open(textfilename, 'w')
text_fh.write(opts.axis_category + '\tdistance\tSampleID' + '\n')
colorby_cats.sort()
plt.figure()
for cat_num, cat in enumerate(colorby_cats):
# collect the primary and secondary samples within this category
state1_samids, state2_samids = get_sam_ids(map_data, map_header,
opts.colorby, cat, opts.primary_state, opts.secondary_state)
state1_samids =\
list(set(state1_samids).intersection(set(distdict.keys())))
state2_samids =\
list(set(state2_samids).intersection(set(distdict.keys())))
if state1_samids == [] or state2_samids == [] or \
(len(state1_samids) == 1 and state1_samids == state2_samids):
raise RuntimeError("one category of samples didn't have any valid" +
" distances. try eliminating samples from -p or -s, or changing" +
" your mapping file with filter_samples_from_otu_table.py")
# go through dmtx
state1_avg_dists = get_avg_dists(
state1_samids,
state2_samids,
distdict)
# plot
xvals = [float(map_dict[sam][opts.axis_category]) for
sam in state1_samids]
try:
color = plt.cm.jet(cat_num / (len(colorby_cats) - 1))
except ZeroDivisionError: # only one cat
color = 'b'
plt.scatter(xvals, state1_avg_dists, edgecolors=color, alpha=.5,
facecolors='none')
plt.xlabel(opts.axis_category)
plt.ylabel('average distance')
lines = [str(xvals[i]) + '\t' + str(state1_avg_dists[i]) +
'\t' + state1_samids[i] + '\n' for i in range(len(xvals))]
text_fh.writelines(lines)
if opts.colorby is not None:
plt.legend(colorby_cats)
plt.savefig(opts.output_path)
def test_get_sam_ids(self):
"""set of sample ids in get_sam_ids should be correct"""
map_file = StringIO.StringIO("""#SampleID Country AgeYears Family AgeCat
h208A.1 Malawi 0.032854209 h208 Child
h301A.1 Malawi 0.05 h301 Child
h301B.1 Malawi 0.05 h301 Child
USinfTw20.1 USA 0.083333333 USinfTw20 Child
USinfTw20.2 USA 0.083333333 USinfTw20 Child
USinfTw1.1 USA 0.083333333 USinfTw1 Child
h10M Malawi 26 h10 Adult
h68M Malawi 26 h68 Adult
TS25 USA 26 USts9 Adult
TS26 USA 26 USts9 Adult""")
map_data, map_header, comments = parse_mapping_file(map_file)
colorby = 'Country'
cat = 'USA'
primary_state = 'AgeCat:Child'
ids1, ids2 = get_sam_ids(map_data, map_header, colorby, cat,
primary_state, secondary_state=None)
self.assertEqual(set(ids1),
set(['USinfTw20.1','USinfTw20.2','USinfTw1.1']))
self.assertEqual(set(ids2), set(['TS25','TS26']))
