def plot_between_subjects_box_plot(metric, where, filename, table='reliability2010_ggmm_thresholding'):
conn = sqlite3.connect(dbfile)
c = conn.cursor()
metric_vals = []
labels = []
# exclude = []
# exclude_str = ""
#
for task in tasks:
for contrast in getStatLabels(task):
where_keys = " and ".join([t[0] + "=?" for t in where+[["task_name"], ["contrast_name"]]])
# exclude = []
# exclude_str = ""
#
# for key in [t[1] for t in where if t[0] == "contrast_name"] + [compare_c]:
# print key
# if key.lower().find('finger') != -1 or key.lower().find('foot') != -1:
# exclude = lefties
# break
#
# if exclude:
# exclude_str = " and (subject_id1 not in (%s) and subject_id2 not in (%s))"%(",".join(["'%s'"%s for s in exclude]), ",".join(["'%s'"%s for s in exclude]))
# print exclude_str
query_str = "SELECT %s FROM %s where %s ORDER BY session, subject_id"%(metric,table,where_keys)
print query_str
c.execute(query_str, tuple([t[1] for t in where] + [task, contrast]))
arr = np.array(c.fetchall())
print arr.shape
metric_vals.append(arr)
labels.append(task+contrast)
if not os.path.exists(os.path.dirname(filename)):
os.makedirs(os.path.dirname(filename))
c.execute("SELECT DISTINCT subject_id FROM reliability2010_ggmm_thresholding ORDER BY subject_id")
subject_ids = c.fetchall()
conn.commit()
c.close()
metric_vals = np.array(metric_vals)
metric_vals.shape = (metric_vals.shape[0], metric_vals.shape[1])
fig = plt.figure(figsize=(8.27,11.69))
figtitle = ", ".join([t[0] + "=" + str(t[1]) for t in where])
t = fig.text(0.5, 0.95, figtitle,
horizontalalignment='center',
fontproperties=mpl.font_manager.FontProperties(size=16))
ax1 = fig.add_subplot(212)
ax1.boxplot(metric_vals.T)
ax1.set_xticklabels(labels)
fig.autofmt_xdate()
ax1.set_ylabel(metric)
#ax1.set_ylim(bottom=-0.1, top=1.1)
subject_ids =[subject_id[0].split('-')[0] for subject_id in subject_ids]
ax2 = fig.add_subplot(211)
colormap = plt.cm.spectral
ax2.set_color_cycle([colormap(i) for i in np.linspace(0, 0.9, len(subject_ids))])
ax2.plot(range(1,len(labels)+1), metric_vals, marker = 'o')
ax2.set_ylabel(metric)
ax2.set_xlim(ax1.get_xlim())
ax2.set_ylim(ax1.get_ylim())
ax2.set_ylim(top=ax2.get_ylim()[1]+3)
ax2.get_xaxis().set_visible(False)
ax2.legend(subject_ids, title="Subjects",
#bbox_to_anchor=(0., 1.02, 1., .102), loc=9,
ncol=4,
mode="expand",
borderaxespad=0.)
if not os.path.exists(os.path.dirname(filename)):
os.makedirs(os.path.dirname(filename))
plt.savefig(filename)
plt.show()
os.makedirs(os.path.dirname(filename))
ax1.text(0.5, 0.90, "onesided_p = %.2f, twosided_p = %.2f"%two_sample(metric_vals[1], metric_vals[0]),
horizontalalignment='center',
fontproperties=mpl.font_manager.FontProperties(size=16),
transform=ax1.transAxes)
plt.savefig(filename)
#plot_between_session_box_plot(task="motor", contrast="Finger", metric="dice", thr_methods=['topo_ggmm', 'topo_fdr'], filename = os.path.join(results_dir, "motor","%s_dice.pdf"%("Finger")))
#compare topo_ggmm with topo_fdr
for roi in rois:
for task in tasks:
for contrast in getStatLabels(task):
if roi:
masked_comparisons = [True]
else:
masked_comparisons = [False, True]
for masked_comparison in masked_comparisons:
plot_between_session_box_plot(where=[('task_name',task), ('contrast_name',contrast), ('roi', roi), ('masked_comparison', masked_comparison)],
metric="dice",
compare= ('thr_method',['topo_ggmm', 'topo_fdr']),
filename = os.path.join(results_dir, "compare topo_ggmm with topo_fdr", "%s_%s_roi_%s_masked_%s_dice.pdf"%(task,contrast, roi, masked_comparison)))
# plot_between_session_box_plot(where=[('task_name',task), ('contrast_name',contrast), ('roi', roi)],
# metric="jaccard",
# compare= ('thr_method',['topo_ggmm', 'topo_fdr']),
# filename = os.path.join(results_dir, "compare topo_ggmm with topo_fdr", "%s_%s_roi_%s_jaccard.pdf"%(task,contrast, roi)))
for roi in rois:
for task in tasks:
