def test_mergeROIs(self):
roi1 = ia.WeightedROI(np.arange(9).reshape((3, 3)))
roi2 = ia.WeightedROI(np.arange(1, 10).reshape((3, 3)))
merged_ROI = sca.merge_weighted_rois(roi1, roi2)
merged_ROI2 = sca.merge_binary_rois(roi1, roi2)
assert (np.array_equal(merged_ROI.get_weighted_mask(),
np.arange(1, 18, 2).reshape((3, 3))))
assert (np.array_equal(merged_ROI2.get_binary_mask(), np.ones((3, 3))))
def test_SpatialReceptiveField_reverse_locations(self):
alts = [-2., -1., 0., 1., 2.]
azis = [-2., -1., 0., 1., 2.]
map = np.array([[0., 0., 0., 0., 0.], [0., 0., 0., 0., 0.],
[0., 0., 1., 1., 0.], [0., 0., 1., 1., 0.],
[0., 0., 0., 0., 0.]])
srf = sca.SpatialReceptiveField(mask=map,
altPos=alts,
aziPos=azis,
sign='ON')
# print(srf.get_weighted_rf_center())
assert (srf.get_weighted_rf_center() == [0.5, 0.5])
srf2 = sca.SpatialReceptiveField(mask=map,
altPos=alts[::-1],
aziPos=azis,
sign='ON')
assert (srf2.get_weighted_rf_center() == [-0.5, 0.5])
srf_f = srf.gaussian_filter(sigma=1)
# print(srf_f.get_weighted_rf_center())
srf2_f = srf2.gaussian_filter(sigma=1)
# print(srf2_f.get_weighted_rf_center())
assert (abs(srf_f.get_weighted_rf_center()[0] +
srf2_f.get_weighted_rf_center()[0]) < 1e-10)
srf_fi = srf_f.interpolate(ratio=2, method='linear')
# print(srf_fi.get_weighted_mask())
# print(srf_fi.get_weighted_rf_center())
# import matplotlib.pyplot as plt
# f=plt.figure()
# ax=f.add_subplot(111)
# srf_fi.plot_rf(plot_axis=ax, tick_spacing=1)
# plt.show()
srf_fit = srf_fi.threshold(thr=0.4)
# print(srf_fit.get_weighted_rf_center())
assert (srf_fit.get_weighted_rf_center()[0] - 0.5 < 0.05)
srf2_fi = srf2_f.interpolate(ratio=2, method='linear')
# print(srf2_fi.get_weighted_mask())
# print(srf2_fi.get_weighted_rf_center())
assert (np.array_equal(srf_fi.get_weighted_mask(),
srf2_fi.get_weighted_mask()))
assert (abs(srf_fi.get_weighted_rf_center()[0] +
srf2_fi.get_weighted_rf_center()[0]) < 1e-10)
def test_SpatialTemporalReceptiveField_plot_traces(self):
locations = [
[2., 0.],
[2., 1.],
[0., 0.],
[0., 1.],
[2., 0.],
[2., 1.],
[0., 0.],
[0., 1.],
]
signs = [1, 1, 1, 1, -1, -1, -1, -1]
traces = [
np.array([[0., 1., 2.]]),
np.array([[2., 1., 0.]]),
np.array([[1., 0., 2.]]),
np.array([[2., 0., 1.]]),
np.array([[0., 0., 2.]]),
np.array([[0., 0., 1.]]),
np.array([[2., 0., 0.]]),
np.array([[1., 0., 0.]]),
]
time = [-1., 0., 1.]
strf = sca.SpatialTemporalReceptiveField(locations=locations,
signs=signs,
traces=traces,
time=time)
strf.plot_traces(figSize=(5, 5), yRange=[-1, 3])
import matplotlib.pyplot as plt
plt.show()
def test_get_local_similarity_index(self):
mask1 = np.zeros((5, 5))
mask1[1:4, 3] = 1
# print(mask1)
mask2 = mask1 * 3
mask3 = mask1 * (-4)
mask4 = np.zeros((5, 5))
mask4[1, 2] = 2
mask4[2, 1] = 5
# print(sca.get_local_similarity_index(mask1, mask2))
#
# print(sca.get_local_similarity_index(mask1, mask3))
#
# print(sca.get_local_similarity_index(mask1, mask4))
assert (np.allclose(sca.get_local_similarity_index(mask1, mask2), 1))
assert (np.allclose(sca.get_local_similarity_index(mask1, mask3), -1))
assert (np.allclose(sca.get_local_similarity_index(mask1, mask4), 0))
def test_DirectionTuning_get_opposite_ind(self):
dire = np.linspace(0, 360, 12, endpoint=False)
resp = np.array(list(range(8)) + list(range(4)), dtype=np.float)
dt = pd.DataFrame(dire, columns=['dire'])
dt['resp_mean'] = resp
# print(dt)
# print(dt.columns)
# print('dire' in dt.columns)
dt = sca.DirectionTuning(data=dt)
print(dt)
assert (dt.get_opposite_ind() == 1)
assert (dt.get_opposite_ind(dire=30) == 7)
assert (dt.get_opposite_ind(dire=225) == 1)
dfe = dt.elevate(bias=5)
print(dfe.trace_type)
def test_SpatialTemporalReceptiveField_IO(self):
locations = [[3.0, 4.0], [3.0, 5.0], [2.0, 4.0], [2.0, 5.0],
[3.0, 4.0], [3.0, 5.0], [2.0, 4.0], [2.0, 5.0]]
signs = [1, 1, 1, 1, -1, -1, -1, -1]
traces = [[np.arange(4)], [np.arange(1, 5)], [np.arange(2, 6)],
[np.arange(3, 7)], [np.arange(5, 9)], [np.arange(6, 10)],
[np.arange(7, 11)], [np.arange(8, 12)]]
time = np.arange(4, 8)
STRF = sca.SpatialTemporalReceptiveField(locations, signs, traces,
time)
# print(STRF.data)
if os.path.isfile(self.testH5Path):
os.remove(self.testH5Path)
testFile = h5py.File(self.testH5Path, 'a')
STRFGroup = testFile.create_group('spatial_temporal_receptive_field')
STRF.to_h5_group(STRFGroup)
testFile.close()
h5File = h5py.File(self.testH5Path, 'r')
STRF2 = sca.SpatialTemporalReceptiveField.from_h5_group(
h5File['spatial_temporal_receptive_field'])
h5File.close()
assert (STRF2.data.altitude.equals(STRF.data.altitude))
assert (STRF2.data.azimuth.equals(STRF.data.azimuth))
assert (STRF2.data.sign.equals(STRF.data.sign))
# print(STRF.data.traces)
# print(STRF2.data.traces)
assert (np.array_equal(
np.array([np.array(t) for t in STRF.data.traces]),
np.array([np.array(t) for t in STRF2.data.traces])))
def test_SpatialTemporalReceptiveField(self):
locations = [[3.0, 4.0], [3.0, 5.0], [2.0, 4.0], [2.0, 5.0],
[3.0, 4.0], [3.0, 5.0], [2.0, 4.0], [2.0, 5.0]]
signs = [1, 1, 1, 1, -1, -1, -1, -1]
traces = [[np.arange(4)], [np.arange(1, 5)], [np.arange(2, 6)],
[np.arange(3, 7)], [np.arange(5, 9)], [np.arange(6, 10)],
[np.arange(7, 11)], [np.arange(8, 12)]]
traces = [np.array(t) for t in traces]
time = np.arange(4, 8)
STRF = sca.SpatialTemporalReceptiveField(locations, signs, traces,
time)
# print(signs)
# print(locations)
# print(traces)
# print(STRF.data)
assert (STRF.data['traces'].iloc[0][0, 1] == 8)
assert (STRF.data['sign'].loc[4] == -1)
newLocations = [[location[0] + 1, location[1] + 1]
for location in locations[0:4]]
newSigns = [1, 1, 1, 1]
STRF.add_traces(newLocations, newSigns, traces[0:4])
assert (STRF.data['traces'][7][1][2] == 4)
def test_getSparseNoiseOnsetIndex(self):
display_log = ft.loadFile(self.sparseNoiseDisplayLogPath)
# print(display_log.keys())
allOnsetInd, onsetIndWithLocationSign = sca.get_sparse_noise_onset_index(
display_log)
# print(list(allOnsetInd[0:10]))
# print(onsetIndWithLocationSign[2])
assert (list(
allOnsetInd[0:10]) == [0, 6, 12, 18, 24, 30, 36, 42, 48, 54])
for probe in onsetIndWithLocationSign:
if np.array_equal(probe[0], np.array([0., 70.
])) and probe[1] == -1.:
# print(probe[2])
assert (np.array_equal(
probe[2],
np.array([
960, 1158, 2280, 3816, 4578, 5586, 6546, 7008, 8496,
9270
])))
def test_SpatialReceptiveField_interpolate(self):
SRF = sca.SpatialReceptiveField(np.random.rand(5, 5),
np.arange(5)[::-1], np.arange(5))
SRF_i = SRF.interpolate(5)
assert (SRF_i.get_weighted_mask().shape == (20, 20))
def test_SpatialReceptiveField_thresholdReceptiveField(self):
SRF = sca.SpatialReceptiveField(
np.arange(9).reshape((3, 3)), np.arange(3), np.arange(3))
thresholdedSRF = SRF.threshold(thr=4)
assert (np.array_equal(thresholdedSRF.weights, np.arange(4, 9)))
def test_SpatialReceptiveField(self):
SRF = sca.SpatialReceptiveField(
np.arange(9).reshape((3, 3)), np.arange(3), np.arange(3))
assert (np.array_equal(SRF.weights, np.arange(1, 9)))
os_diff_on = []
os_diff_off = []
for roi_i, roi_row in osdf.iterrows():
date = int(roi_row['date'])
mid = roi_row['mouse_id']
plane_n = roi_row['plane_n']
roi_n = roi_row['roi_n']
map_fn = '{}_{}_{}_{}'.format(date, mid, plane_n, response_dir)
map_f = h5py.File(os.path.join(rf_maps_folder, map_fn + '.hdf5'), 'r')
on_grp = map_f['{}_ON'.format(map_fn)]
off_grp = map_f['{}_OFF'.format(map_fn)]
dire = roi_row['dgc_{}_{}_{}_{}'.format(response_dir, dire_type, dire_pp, response_type)]
ori = sca.dire2ori(dire)
if roi_n in on_grp.keys() and roi_n in off_grp.keys():
rf_on = sca.SpatialReceptiveField.from_h5_group(on_grp[roi_n])
rf_off = sca.SpatialReceptiveField.from_h5_group(off_grp[roi_n])
c_alt_on, c_azi_on = rf_on.get_weighted_rf_center()
c_alt_off, c_azi_off = rf_off.get_weighted_rf_center()
onoff_ang = np.arctan((c_alt_on - c_alt_off) / (c_azi_on - c_azi_off))
onoff_ang = onoff_ang * 180. / np.pi
onoff_ang = sca.dire2ori(onoff_ang)
curr_diff = abs(onoff_ang - ori)
if curr_diff > 90.:
curr_diff = 180 - curr_diff
pdff = PdfPages(os.path.join(save_folder, 'STRFs_' + plane_n + '.pdf'))
roi_lst = nwb_f['processing/rois_and_traces_' + plane_n + '/ImageSegmentation/imaging_plane/roi_list'].value
roi_lst = [r for r in roi_lst if r[:4] == 'roi_']
roi_lst.sort()
for roi_ind, roi_n in enumerate(roi_lst):
print('roi: {} / {}'.format(roi_ind + 1, len(roi_lst)))
curr_trace, _ = dt.get_single_trace(nwb_f=nwb_f, plane_n=plane_n, roi_n=roi_n)
if np.min(curr_trace) < bias:
add_to_trace = -np.min(curr_trace) + bias
else:
add_to_trace = 0.
curr_strf = sca.get_strf_from_nwb(h5_grp=strf_grp[plane_n], roi_ind=roi_ind, trace_type=trace_type)
curr_strf_dff = curr_strf.get_local_dff_strf(is_collaps_before_normalize=True, add_to_trace=add_to_trace)
v_min, v_max = curr_strf_dff.get_data_range()
f = curr_strf_dff.plot_traces(yRange=(v_min, v_max * 1.1), figSize=(16, 10),
columnSpacing=0.002, rowSpacing=0.002)
# plt.show()
pdff.savefig(f)
f.clear()
plt.close(f)
pdff.close()
nwb_f.close()
_, rf_off_new = dt.get_rf_properties(
srf=srf_off,
polarity='negative',
sigma=analysis_params['gaussian_filter_sigma_rf'],
interpolate_rate=analysis_params[
'interpolate_rate_rf'],
z_thr_abs=analysis_params['rf_z_thr_abs'],
z_thr_rel=analysis_params['rf_z_thr_rel'])
else:
raise ValueError
rf_on_mask = rf_on_new.get_weighted_mask()
rf_off_mask = rf_off_new.get_weighted_mask()
rf_onoff_new = sca.SpatialReceptiveField(
mask=np.max([rf_on_mask, rf_off_mask], axis=0),
altPos=rf_on_new.altPos,
aziPos=rf_on_new.aziPos,
sign='ON_OFF',
thr=analysis_params['rf_z_thr_abs'])
curr_s2_grp = s2_grp.create_group(roi_row['roi_n'])
rf_onoff_new.to_h5_group(curr_s2_grp)
curr_s1_on_grp = s1_on_grp.create_group(roi_row['roi_n'])
rf_on_new.to_h5_group(curr_s1_on_grp)
curr_s1_off_grp = s1_off_grp.create_group(roi_row['roi_n'])
rf_off_new.to_h5_group(curr_s1_off_grp)
# positive S1 ON
s1_on_df = subdf[(subdf['rf_{}_on_peak_z'.format(response_dir)] >=
analysis_params['rf_z_thr_abs'])