Skip to content

Commit 9877b65

Browse files
mayofaulknermayofaulkner
committed
quickish way to gwt all channels
1 parent 04a2135 commit 9877b65

File tree

3 files changed

+318
-19
lines changed

3 files changed

+318
-19
lines changed

‎needles2/needles_viewer.py‎

Lines changed: 14 additions & 2 deletions
Original file line numberDiff line numberDiff line change
@@ -15,13 +15,13 @@ class NeedlesViewer(AtlasController):
1515
def __init__(self):
1616
super(NeedlesViewer, self).__init__()
1717

18-
def initialize(self, plot, resolution=25, mapping='Allen', num_windows=1, render=False):
18+
def initialize(self, plot, resolution=25, mapping='Allen', volume_mode=None, num_windows=1, render=False):
1919
vedo.settings.allowInteraction = False
2020
self.plot = plot
2121
self.plot_window_id = 0
2222
self.model = AtlasModel()
2323
self.model.initialize(resolution)
24-
self.model.load_allen_volume(mapping)
24+
self.model.load_allen_volume(mapping, volume_mode)
2525
self.model.initialize_slicers()
2626

2727
self.view = AtlasView(self.plot, self.model)
@@ -72,4 +72,16 @@ def initialize_embed_ui(self, slicer_target=None):
7272
#self.add_slider('pz', self.update_pz_slicer, 0, int(d[2]), 0, pos=(0.35, 0.065, 0.12),
7373
# title='+Z', **s_kw)
7474

75+
def update_slicer(self, slicer_view, value):
76+
"""
77+
Update a given slicer with the given value
78+
:param slicer_view: SlicerView instance
79+
:param value: Value
80+
"""
81+
volume = self.view.volume
82+
model = slicer_view.model
83+
model.set_value(value)
84+
model.clipping_planes = volume.get_clipping_planes(model.axis)
85+
slicer_view.update(add_to_scene=self.model.slices_visible)
86+
7587

‎needles2/probe_model.py‎

Lines changed: 191 additions & 0 deletions
Original file line numberDiff line numberDiff line change
@@ -0,0 +1,191 @@
1+
import numpy as np
2+
from brainbox.numerical import ismember
3+
from oneibl.one import ONE
4+
from ibllib.pipes import histology
5+
from ibllib.atlas import AllenAtlas
6+
from ibllib.ephys.neuropixel import TIP_SIZE_UM, SITES_COORDINATES
7+
from ibllib.pipes.ephys_alignment import EphysAlignment
8+
import time
9+
10+
PROV_2_VAL = {
11+
'Resolved': 90,
12+
'Ephys aligned histology track': 70,
13+
'Histology track': 50,
14+
'Micro-manipulator': 30,
15+
'Planned': 10}
16+
17+
VAL_2_PROV = {v: k for k, v in PROV_2_VAL.items()}
18+
19+
20+
class ProbeModel:
21+
def __init__(self, one=None, ba=None):
22+
23+
self.one = one or ONE()
24+
self.ba = ba or AllenAtlas(25)
25+
self.traj = {'Planned': {},
26+
'Micro-manipulator': {},
27+
'Histology track': {},
28+
'Ephys aligned histology track': {},
29+
'Resolved': {},
30+
'Best': {}}
31+
32+
self.get_traj_for_provenance(provenance='Histology track', django=['x__isnull,False'])
33+
self.get_traj_for_provenance(provenance='Ephys aligned histology track')
34+
self.get_traj_for_provenance(provenance='Ephys aligned histology track',
35+
django=['probe_insertion__json__extended_qc__'
36+
'alignment_resolved,True'], prov_dict='Resolved')
37+
self.find_traj_is_best(provenance='Histology track')
38+
self.find_traj_is_best(provenance='Ephys aligned histology track')
39+
self.traj['Resolved']['is_best'] = np.arange(len(self.traj['Resolved']['traj']))
40+
41+
@staticmethod
42+
def get_traj_info(traj):
43+
return traj['probe_insertion'], traj['x'], traj['y']
44+
45+
def get_traj_for_provenance(self, provenance='Histology track', django=None,
46+
prov_dict=None):
47+
48+
if django is None:
49+
django = []
50+
51+
if prov_dict is None:
52+
prov_dict = provenance
53+
54+
django_base = ['probe_insertion__session__project__name__icontains,'
55+
'ibl_neuropixel_brainwide_01']
56+
django_str = ','.join(django_base + django)
57+
58+
self.traj[prov_dict]['traj'] = np.array(self.one.alyx.rest('trajectories', 'list',
59+
provenance=provenance,
60+
django=django_str))
61+
ins_ids, x, y = zip(*[self.get_traj_info(traj) for traj in self.traj[prov_dict]['traj']])
62+
self.traj[prov_dict]['ins'] = np.array(ins_ids)
63+
self.traj[prov_dict]['x'] = np.array(x)
64+
self.traj[prov_dict]['y'] = np.array(y)
65+
66+
def compute_best_for_provenance(self, provenance='Histology track'):
67+
val = PROV_2_VAL[provenance]
68+
prov_to_include = []
69+
for k, v in VAL_2_PROV.items():
70+
if k >= val:
71+
prov_to_include.append(v)
72+
73+
for iP, prov in enumerate(prov_to_include):
74+
if not 'is_best' in self.traj[prov].keys():
75+
self.find_traj_is_best(prov)
76+
77+
if iP == 0:
78+
self.traj['Best']['traj'] = self.traj[prov]['traj'][self.traj[prov]['is_best']]
79+
self.traj['Best']['ins'] = self.traj[prov]['ins'][self.traj[prov]['is_best']]
80+
self.traj['Best']['x'] = self.traj[prov]['x'][self.traj[prov]['is_best']]
81+
self.traj['Best']['y'] = self.traj[prov]['y'][self.traj[prov]['is_best']]
82+
else:
83+
self.traj['Best']['traj'] = np.r_[self.traj['Best']['traj'],
84+
(self.traj[prov]['traj']
85+
[self.traj[prov]['is_best']])]
86+
self.traj['Best']['ins'] = np.r_[self.traj['Best']['ins'],
87+
(self.traj[prov]['ins']
88+
[self.traj[prov]['is_best']])]
89+
self.traj['Best']['x'] = np.r_[self.traj['Best']['x'],
90+
self.traj[prov]['x'][self.traj[prov]['is_best']]]
91+
self.traj['Best']['y'] = np.r_[self.traj['Best']['y'],
92+
self.traj[prov]['y'][self.traj[prov]['is_best']]]
93+
94+
def find_traj_is_best(self, provenance='Histology track'):
95+
val = PROV_2_VAL[provenance]
96+
next_provenance = VAL_2_PROV[val + 20]
97+
98+
if not 'traj' in self.traj[provenance].keys():
99+
self.get_traj_for_provenance(provenance)
100+
if not 'traj' in self.traj[next_provenance].keys():
101+
self.get_traj_for_provenance(next_provenance)
102+
103+
isin, _ = ismember(self.traj[provenance]['ins'],
104+
self.traj[next_provenance]['ins'])
105+
self.traj[provenance]['is_best'] = np.where(np.invert(isin))[0]
106+
107+
# Special exception for planned provenance
108+
if provenance == 'Planned':
109+
next_provenance = VAL_2_PROV[val + 40]
110+
if not 'traj' in self.traj[next_provenance].keys():
111+
self.get_traj_for_provenance(next_provenance)
112+
isin, _ = ismember(self.traj[provenance]['ins'][self.traj[provenance]['is_best']],
113+
self.traj[next_provenance]['ins'])
114+
self.traj[provenance]['is_best'] = (self.traj[provenance]['is_best']
115+
[np.where(np.invert(isin))[0]])
116+
117+
def get_channels(self, provenance):
118+
119+
depths = SITES_COORDINATES[:, 1]
120+
start = time.time()
121+
122+
# Need to account for case when no insertions for planned and micro can skip the insertions
123+
insertions = []
124+
step = 150
125+
for i in range(np.ceil(self.traj[provenance]['ins'].shape[0] / step).astype(np.int)):
126+
insertions += self.one.alyx.rest(
127+
'insertions', 'list', django=f"id__in,{list(self.traj[provenance]['ins'][i * step:(i + 1) * step])}")
128+
129+
ins_id = np.ravel([np.where(ins['id'] == self.traj[provenance]['ins'])
130+
for ins in insertions])
131+
132+
end = time.time()
133+
print(end-start)
134+
135+
start1 = time.time()
136+
for iT, (traj, ins) in enumerate(zip(self.traj[provenance]['traj'][ins_id], insertions)):
137+
try:
138+
xyz_picks = np.array(ins['json']['xyz_picks']) / 1e6
139+
if traj['provenance'] == 'Histology track':
140+
xyz_picks = xyz_picks[np.argsort(xyz_picks[:, 2]), :]
141+
xyz_channels = histology.interpolate_along_track(xyz_picks, (depths + TIP_SIZE_UM) / 1e6)
142+
else:
143+
align_key = ins['json']['extended_qc']['alignment_stored']
144+
feature = traj['json'][align_key][0]
145+
track = traj['json'][align_key][1]
146+
ephysalign = EphysAlignment(xyz_picks, depths, track_prev=track,
147+
feature_prev=feature,
148+
brain_atlas=self.ba, speedy=True)
149+
xyz_channels = ephysalign.get_channel_locations(feature, track)
150+
151+
if iT == 0:
152+
all_channels = xyz_channels
153+
else:
154+
all_channels = np.r_[all_channels, xyz_channels]
155+
except Exception as err:
156+
print(err)
157+
print(traj['id'])
158+
159+
end = time.time()
160+
print(end-start1)
161+
162+
return all_channels
163+
164+
from scipy.signal import fftconvolve
165+
cvol = np.zeros(ba.image.shape, dtype=np.float)
166+
val, counts = np.unique(ba._lookup(all_channels), return_counts=True)
167+
cvol[np.unravel_index(val, cvol.shape)] = counts
168+
#
169+
from ibllib.dsp import fcn_cosine
170+
DIST_FCN = np.array([100, 150]) / 1e6
171+
dx = ba.bc.dx
172+
template = np.arange(- np.max(DIST_FCN) - dx, np.max(DIST_FCN) + 2 * dx, dx) ** 2
173+
kernel = sum(np.meshgrid(template, template, template))
174+
kernel = 1 - fcn_cosine(DIST_FCN)(np.sqrt(kernel))
175+
#
176+
cvol = fftconvolve(cvol, kernel)
177+
#
178+
#
179+
# cvol[np.unravel_index(ba._lookup(all_channels), cvol.shape)] = 1
180+
181+
# from ibllib.atlas import AllenAtlas
182+
# ba = AllenAtlas()
183+
# import vedo
184+
# import numpy as np
185+
#
186+
# actor = vedo.Volume(ba.image, c='bone', spacing = np.array([25]*3), mapper='smart', mode=0, alphaGradient=0.5)
187+
# plt = vedo.Plotter()
188+
# plt.add(actor)
189+
# plt.show()
190+
191+

0 commit comments

Comments
 (0)