-
Notifications
You must be signed in to change notification settings - Fork 0
/
SI_sorting.py
1311 lines (1075 loc) · 73.4 KB
/
SI_sorting.py
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
61
62
63
64
65
66
67
68
69
70
71
72
73
74
75
76
77
78
79
80
81
82
83
84
85
86
87
88
89
90
91
92
93
94
95
96
97
98
99
100
101
102
103
104
105
106
107
108
109
110
111
112
113
114
115
116
117
118
119
120
121
122
123
124
125
126
127
128
129
130
131
132
133
134
135
136
137
138
139
140
141
142
143
144
145
146
147
148
149
150
151
152
153
154
155
156
157
158
159
160
161
162
163
164
165
166
167
168
169
170
171
172
173
174
175
176
177
178
179
180
181
182
183
184
185
186
187
188
189
190
191
192
193
194
195
196
197
198
199
200
201
202
203
204
205
206
207
208
209
210
211
212
213
214
215
216
217
218
219
220
221
222
223
224
225
226
227
228
229
230
231
232
233
234
235
236
237
238
239
240
241
242
243
244
245
246
247
248
249
250
251
252
253
254
255
256
257
258
259
260
261
262
263
264
265
266
267
268
269
270
271
272
273
274
275
276
277
278
279
280
281
282
283
284
285
286
287
288
289
290
291
292
293
294
295
296
297
298
299
300
301
302
303
304
305
306
307
308
309
310
311
312
313
314
315
316
317
318
319
320
321
322
323
324
325
326
327
328
329
330
331
332
333
334
335
336
337
338
339
340
341
342
343
344
345
346
347
348
349
350
351
352
353
354
355
356
357
358
359
360
361
362
363
364
365
366
367
368
369
370
371
372
373
374
375
376
377
378
379
380
381
382
383
384
385
386
387
388
389
390
391
392
393
394
395
396
397
398
399
400
401
402
403
404
405
406
407
408
409
410
411
412
413
414
415
416
417
418
419
420
421
422
423
424
425
426
427
428
429
430
431
432
433
434
435
436
437
438
439
440
441
442
443
444
445
446
447
448
449
450
451
452
453
454
455
456
457
458
459
460
461
462
463
464
465
466
467
468
469
470
471
472
473
474
475
476
477
478
479
480
481
482
483
484
485
486
487
488
489
490
491
492
493
494
495
496
497
498
499
500
501
502
503
504
505
506
507
508
509
510
511
512
513
514
515
516
517
518
519
520
521
522
523
524
525
526
527
528
529
530
531
532
533
534
535
536
537
538
539
540
541
542
543
544
545
546
547
548
549
550
551
552
553
554
555
556
557
558
559
560
561
562
563
564
565
566
567
568
569
570
571
572
573
574
575
576
577
578
579
580
581
582
583
584
585
586
587
588
589
590
591
592
593
594
595
596
597
598
599
600
601
602
603
604
605
606
607
608
609
610
611
612
613
614
615
616
617
618
619
620
621
622
623
624
625
626
627
628
629
630
631
632
633
634
635
636
637
638
639
640
641
642
643
644
645
646
647
648
649
650
651
652
653
654
655
656
657
658
659
660
661
662
663
664
665
666
667
668
669
670
671
672
673
674
675
676
677
678
679
680
681
682
683
684
685
686
687
688
689
690
691
692
693
694
695
696
697
698
699
700
701
702
703
704
705
706
707
708
709
710
711
712
713
714
715
716
717
718
719
720
721
722
723
724
725
726
727
728
729
730
731
732
733
734
735
736
737
738
739
740
741
742
743
744
745
746
747
748
749
750
751
752
753
754
755
756
757
758
759
760
761
762
763
764
765
766
767
768
769
770
771
772
773
774
775
776
777
778
779
780
781
782
783
784
785
786
787
788
789
790
791
792
793
794
795
796
797
798
799
800
801
802
803
804
805
806
807
808
809
810
811
812
813
814
815
816
817
818
819
820
821
822
823
824
825
826
827
828
829
830
831
832
833
834
835
836
837
838
839
840
841
842
843
844
845
846
847
848
849
850
851
852
853
854
855
856
857
858
859
860
861
862
863
864
865
866
867
868
869
870
871
872
873
874
875
876
877
878
879
880
881
882
883
884
885
886
887
888
889
890
891
892
893
894
895
896
897
898
899
900
901
902
903
904
905
906
907
908
909
910
911
912
913
914
915
916
917
918
919
920
921
922
923
924
925
926
927
928
929
930
931
932
933
934
935
936
937
938
939
940
941
942
943
944
945
946
947
948
949
950
951
952
953
954
955
956
957
958
959
960
961
962
963
964
965
966
967
968
969
970
971
972
973
974
975
976
977
978
979
980
981
982
983
984
985
986
987
988
989
990
991
992
993
994
995
996
997
998
999
1000
import spikeinterface.core as si
import spikeinterface.preprocessing as spre
from probeinterface.plotting import plot_probe
from spikeinterface.extractors import read_nwb
from spikeinterface.sortingcomponents.peak_detection import detect_peaks
from spikeinterface.sortingcomponents.peak_localization import localize_peaks
from spikeinterface.sortingcomponents.motion import estimate_motion, interpolate_motion
from spikeinterface.sorters import run_sorter, read_sorter_folder
import spikeinterface.curation as scur
import spikeinterface.qualitymetrics as sqm
from spikeinterface.exporters import export_to_phy
import matplotlib.pyplot as plt
import SI_tools
import os
import shutil
import numpy
import pickle
####################################################################################################################################################################################
# SOME DEFAULT PARAMETERS
####################################################################################################################################################################################
n_cpus = os.cpu_count()
n_jobs = n_cpus - 2
job_kwargs = dict(chunk_duration="1s", n_jobs=n_jobs, progress_bar=True)
# si.set_global_job_kwargs(**job_kwargs)
######################################################################################################
# PARAMETER FOR "LOCAL REFERENCING"
exclude_radius_chans_default = 1 # Number of neighbor channels to exclude because are too close to the reference channel
include_radius_chans_default = 4 # Number of neighbor channels delineates the outer boundary of the annulus whose role is to exclude channels that are too far away
noisy_freq_default = None
######################################################################################################
# GENERAL PARAMETER FOR "PEAKS LOCATIONS"
ms_before_default = 0.6
ms_after_default = 1.5
peak_sign_default = 'both' # (“neg” | “pos” | “both”)
nearest_chans_default = 3 # Number of neighbor channels to search for the same waveform
###########################################################
# DETECT PEAKS:
peak_detect_threshold_default = 5 # MAD: Median Amplitude Deviations
######################################################################################################
# MOTION ESTIMATION & INTERPOLATION
# monopolar triangulation + non-rigid + decentralized
motion_rigid_default = False
motion_options_default = {
'method': 'dredge_ap', # Paninski Lab
'method_kwargs' : {}
}
interpolate_options_default = {
'method': 'kriging', # Kilosort-like
'border_mode': 'remove_channels' # ('remove_channels' | 'force_extrapolate' | 'force_zeros')
}
sorterName_default = 'kilosort4'
####################################################################################################################################################################################
####################################################################################################################################################################################
# MAIN FUNCTIONS
####################################################################################################################################################################################
####################################################################################################################################################################################
####################################################################################################################################################################################
# Helper function to search for unique *NWB files from expDATES within a range of dates
####################################################################################################################################################################################
def get_expDay_in_range(parentRecordingFolder, year_start, month_start, day_start, year_stop, month_stop, day_stop):
filesDate_log = []
filesData_numList = []
for _, _, files in os.walk(parentRecordingFolder):
for name in files:
nameSplit = os.path.splitext(name)
if nameSplit[1]=='.nwb' and '-noNEV' not in nameSplit[0]:
fileName = nameSplit[0]
yearFile = int(fileName[3:7])
monthFile = int(fileName[8:10])
dayFile = int(fileName[11:13])
fileIN = True
if yearFile>=year_start and yearFile<=year_stop:
if yearFile==year_start:
if monthFile<month_start:
fileIN = False
elif monthFile==month_start and dayFile<day_start:
fileIN = False
elif yearFile==year_stop:
if monthFile>month_stop:
fileIN = False
elif monthFile==month_stop and dayFile>day_stop:
fileIN = False
if fileIN:
file_label = '{}-{:02d}-{:02d}'.format(yearFile, monthFile, dayFile)
if file_label not in filesDate_log:
filesDate_log.append(file_label)
filesData_numList.append([yearFile, monthFile, dayFile])
# Force Unique and increasing
dateSort = numpy.unique(numpy.array(filesData_numList), axis=0)
return dateSort
####################################################################################################################################################################################
# RUN PREPROCESSING for all the Electrode/Probes and sessions from expDays within a range of dates
####################################################################################################################################################################################
def run_prepro_expDAY_in_range(parentRecordingFolder, parentPreproFolder,
year_start, month_start, day_start,
year_stop, month_stop, day_stop,
local_radius_chans = (exclude_radius_chans_default, include_radius_chans_default),
noisy_freq = noisy_freq_default,
ms_before = ms_before_default,
ms_after = ms_after_default,
peak_sign = peak_sign_default,
nearest_chans = nearest_chans_default,
peak_detect_threshold = peak_detect_threshold_default,
do_motion = True,
motion_rigid = motion_rigid_default,
motion_options = motion_options_default,
interpolate_options = interpolate_options_default,
localProcess_NWB = False,
rewrite_prepro = True
):
dateSort = get_expDay_in_range(parentRecordingFolder, year_start, month_start, day_start, year_stop, month_stop, day_stop)
for n in range(dateSort.shape[0]):
sessionYear=dateSort[n, 0]
sessionMonth=dateSort[n, 1]
sessionDay=dateSort[n, 2]
run_prepro_expDAY(parentRecordingFolder, parentPreproFolder, sessionYear, sessionMonth, sessionDay,
local_radius_chans = local_radius_chans,
noisy_freq = noisy_freq,
ms_before = ms_before,
ms_after = ms_after,
peak_sign = peak_sign,
nearest_chans = nearest_chans,
peak_detect_threshold = peak_detect_threshold,
do_motion = do_motion,
motion_rigid = motion_rigid,
motion_options = motion_options,
interpolate_options = interpolate_options,
localProcess_NWB = localProcess_NWB,
rewrite_prepro = rewrite_prepro
)
####################################################################################################################################################################################
# RUN PREPROCESSING for all the Electrode/Probes (ALL sessions) for a given expDay
####################################################################################################################################################################################
def run_prepro_expDAY(parentRecordingFolder, parentPreproFolder, sessionYear, sessionMonth, sessionDay,
local_radius_chans = (exclude_radius_chans_default, include_radius_chans_default),
noisy_freq = noisy_freq_default,
ms_before = ms_before_default,
ms_after = ms_after_default,
peak_sign = peak_sign_default,
nearest_chans = nearest_chans_default,
peak_detect_threshold = peak_detect_threshold_default,
do_motion = True,
motion_rigid = motion_rigid_default,
motion_options = motion_options_default,
interpolate_options = interpolate_options_default,
localProcess_NWB = False,
rewrite_prepro = True
):
print('Preprocessing exp{}-{:02d}-{:02d}'.format(sessionYear, sessionMonth, sessionDay))
electrodeGroups = SI_tools.getUnique_electrodeGroups(parentRecordingFolder, sessionYear, sessionMonth, sessionDay)
for i in range(len(electrodeGroups)):
if any([prefix in electrodeGroups[i]['probeInfo']['probeName'] for prefix in SI_tools.supported_probes_manufacturer]):
electrodeGroup_sessions = SI_tools.select_electrodeGroup_and_session_info(electrodeGroups, sessionYear, sessionMonth, sessionDay, electrodeGroup_Name=electrodeGroups[i]['electrodeName'])
run_prepro(parentRecordingFolder, parentPreproFolder, sessionYear, sessionMonth, sessionDay,
electrodeGroup_sessions = electrodeGroup_sessions,
local_radius_chans = local_radius_chans,
noisy_freq = noisy_freq,
ms_before = ms_before,
ms_after = ms_after,
peak_sign = peak_sign,
nearest_chans = nearest_chans,
peak_detect_threshold = peak_detect_threshold,
do_motion = do_motion,
motion_rigid = motion_rigid,
motion_options = motion_options,
interpolate_options = interpolate_options,
localProcess_NWB = localProcess_NWB,
rewrite_prepro = rewrite_prepro,
return_recording = False
)
print('\n\nexp{}-{:02d}-{:02d} ElectrodeGroup : {} was processed ¡¡¡\n\n'.format(sessionYear, sessionMonth, sessionDay, electrodeGroups[i]['electrodeName']))
else:
print('\nProbe: {} was not found as a valid Device\nFrom exp{}-{:02d}-{:02d}, ElectrodeGroup : "{}"\nIt will NOT be preprocessing\n\n'.format(
electrodeGroups[i]['probeInfo']['probeName'], sessionYear, sessionMonth, sessionDay, electrodeGroups[i]['electrodeName']))
####################################################################################################################################################################################
# Function to set up parameters to run Sorter & to create Sorting_analyzer
####################################################################################################################################################################################
def get_sorting_params(sorterName, nChans, step_chan, sampling_frequency, ms_before, ms_after, peak_sign, nearest_chans, sorter_whitening, detect_threshold=None):
# Validate Nearest channels relative to the number of channels in the recording object
#"nearest_chans = 0" will perform similar to sorting single channels.
if nChans==1:
nearest_chans = 0
elif nearest_chans>nChans:
nearest_chans = nChans
if nearest_chans < 1:
radius_um = step_chan/2
location_method = "center_of_mass"
unit_location_kwargs = {
'feature': "ptp" #"ptp" | "mean" | "energy" | "peak_voltage", default: "ptp"
}
else:
radius_um = step_chan*nearest_chans
location_method = "monopolar_triangulation"
unit_location_kwargs = {
'max_distance_um': step_chan*10,
'optimizer': 'least_square'
}
snippet_T1 = int(numpy.ceil(ms_before * sampling_frequency / 1000.0))
snippet_T2 = int(numpy.ceil(ms_after * sampling_frequency / 1000.0))
####################################################################################
# Sparsity will be the first extension to be computed.
estimate_sparsity_params = {
'num_spikes_for_sparsity': 500, # How many spikes per units to compute the sparsity (default: int = 100)
'ms_before': ms_before, # Cut out in ms before spike time (default: float = 1.0)
'ms_after': ms_after, # Cut out in ms after spike time (default: float'= 2.5)
'method': "radius", # ("radius" | "best_channels" | "amplitude" | "snr" | "by_property" | "ptp" (default: str = 'radius')
'peak_sign': peak_sign, # Sign of the template to compute best channels (“neg” | “pos” | “both” (default: str = 'neg')
'radius_um': radius_um, # (default: float = 100.0) Radius in um for “radius” method
'num_channels': nearest_chans, # Used for “best_channels” method (default: int = 5)
}
####################################################################################
# After sparsity is perfomed, then the rest of Postprocessing extensions can be listed:
# All extensions.
# Default parameters are listed as comments
if nearest_chans>=5:
include_multi_channel_metrics = True
metric_names = ['peak_to_valley', 'peak_trough_ratio', 'halfwidth', 'repolarization_slope', 'recovery_slope', 'num_positive_peaks', 'num_negative_peaks',
'velocity_above', 'velocity_below', 'exp_decay', 'spread']
else:
include_multi_channel_metrics = False
metric_names = []
sorting_analyzer_params = {
'random_spikes': {'method': 'uniform', 'max_spikes_per_unit': 500, 'margin_size': None}, # 'method': 'uniform' | 'all', 'max_spikes_per_unit': 500, 'margin_size': None
'noise_levels': {}, # method : 'mad' | 'std', str default = 'mad' # it is not fully integrated to Extensions factory, it doesn't have the function "._set_params()"
'correlograms': {'window_ms' : 50.0, 'bin_ms': 1.0}, # 'window_ms' : 50 (if 50 ms, the correlations will be computed at lags -25 ms … 25 ms), 'bin_ms' : 1
'isi_histograms': {'window_ms' : 50.0, 'bin_ms': 1.0}, # 'window_ms' : 50, 'bin_ms' : 1
'waveforms': {'ms_before': ms_before, 'ms_after': ms_after},
'principal_components': {'n_components': 5, 'mode': 'by_channel_local'}, # 'n_components': 5, 'mode': 'by_channel_local' | by_channel_global, default: by_channel_local
'templates': {'operators': ["average"], 'ms_before': ms_before, 'ms_after': ms_after}, # The operators to compute. Can be "average", "std", "median", "percentile" , 'ms_before': 1, 'ms_after': 2
'template_metrics': {'peak_sign': peak_sign,
'upsampling_factor': 20, # The upsampling factor to upsample the templates, default: 10
'sparsity': None, # If None, template metrics are computed on the extremum channel only. If sparsity is given, template metrics are computed on all sparse channels of each unit. Default: None
'include_multi_channel_metrics': include_multi_channel_metrics, # Whether to compute multi-channel metrics (At least 10 channels shoulb be capturing the waveforms)
'metric_names': metric_names, # ['peak_to_valley', 'peak_trough_ratio', 'halfwidth', 'repolarization_slope', 'recovery_slope', 'num_positive_peaks', 'num_negative_peaks',
# 'velocity_above', 'velocity_below', 'exp_decay', 'spread'], # the following multi-channel metrics can be computed (when include_multi_channel_metrics=True)
'recovery_window_ms': 0.7, # the window in ms after the peak to compute the recovery_slope
'peak_relative_threshold': 0.2, #the relative threshold to detect positive and negative peaks, default: 0.2
'peak_width_ms': 0.1, # the width in samples to detect peaks, default: 0.1
'depth_direction': "y", # the direction to compute velocity above and below, default: "y" (see notes)
'min_channels_for_velocity': 5, #the minimum number of channels above or below to compute velocity, default: 5
'min_r2_velocity': 0.5, # the minimum r2 to accept the velocity fit, default: 0.5
'exp_peak_function': 'ptp', #the function to use to compute the peak amplitude for the exp decay, default: "ptp"
'min_r2_exp_decay': 0.5, # the minimum r2 to accept the exp decay fit, default: 0.5
'spread_threshold': 0.2, # the threshold to compute the spread, default: 0.2
'spread_smooth_um': step_chan, # the smoothing in um to compute the spread, default: 20
'column_range': None, # the range in um in the horizontal direction to consider channels for velocity, default: None, If None, all channels all channels are considered,
# If 0 or 1, only the "column" that includes the max channel is considered
# If > 1, only channels within range (+/-) um from the max channel horizontal position are used
# Notes
# -----
# If any multi-channel metric is in the metric_names or include_multi_channel_metrics is True, sparsity must be None,
# so that one metric value will be computed per unit.
# For multi-channel metrics, 3D channel locations are not supported. By default, the depth direction is "y".
},
'template_similarity': {'method': 'cosine', 'max_lag_ms': 0.0}, # 'method': “cosine” | “l2” | “l1”, 'max_lag_ms': 0.0
'amplitude_scalings': {'ms_before': ms_before, 'ms_after': ms_after, 'handle_collisions': True, 'delta_collision_ms': 2}, # 'handle_collisions': True, delta_collision_ms: 2
'spike_amplitudes': {'peak_sign': peak_sign}, # ( “neg” | “pos” | “both”, default: str = 'neg')
'spike_locations': {
'ms_before': ms_before, # ms_before : 0.5,
'ms_after': ms_after, # ms_after : 0.5,
'method': location_method, # 'method': "center_of_mass" | "monopolar_triangulation" | "grid_convolution", default: 'center_of_mass'
'spike_retriver_kwargs': {
'channel_from_template': False, # For each spike is the maximum channel computed from template or re estimated at every spikes. float, default: 50
'radius_um': radius_um, # In case channel_from_template=False, this is the radius to get the true peak. bool, default = True
'peak_sign': peak_sign, # In case channel_from_template=False, this is the peak sign. ( “neg” | “pos” | “both”, default: str = 'neg')
},
},
'unit_locations': {
'method': location_method, #'method': "center_of_mass" | "monopolar_triangulation" | "grid_convolution", default: 'center_of_mass'
'radius_um': radius_um,
**unit_location_kwargs
}
}
if 'both' in peak_sign:
detect_sign = 0
elif 'neg' in peak_sign:
detect_sign = -1
elif 'pos' in peak_sign:
detect_sign = 1
nt = snippet_T1 + snippet_T2
if nearest_chans>1:
scheme2_detect_channel_radius = radius_um/2
else:
scheme2_detect_channel_radius = radius_um
if sorterName=='mountainsort5':
if detect_threshold is None:
detect_threshold = 5.5 # ZCA-Whitening traces
sorter_label = 'MS5ch{}th{}'.format(nearest_chans, detect_threshold).replace('.', '_')
sorter_info = dict(
sorter_name = 'mountainsort5',
sorter_label = sorter_label,
useDocker = False,
params2update = {
'scheme': '2', # Not searching for templates in chunks
'detect_threshold': detect_threshold, # ZCA-Whitening traces
'detect_sign': detect_sign,
'snippet_T1': snippet_T1,
'snippet_T2': snippet_T2,
'snippet_mask_radius': radius_um,
'scheme1_detect_channel_radius': radius_um,
'scheme2_phase1_detect_channel_radius': radius_um,
'scheme2_detect_channel_radius': scheme2_detect_channel_radius,
'scheme2_training_recording_sampling_mode': 'initial',
'filter': False,
'whiten': sorter_whitening,
**job_kwargs,
},
estimate_sparsity_params = estimate_sparsity_params,
sorting_analyzer_params = sorting_analyzer_params
)
elif sorterName =='kilosort4':
if detect_threshold is None:
detect_threshold = 3 # Whitened Threshold (STD from Absolute traces)
sorter_label = 'KS4ch{}th{}'.format(nearest_chans, detect_threshold).replace('.', '_')
sorter_info= dict(
sorter_name = 'kilosort4',
sorter_label = sorter_label,
useDocker = False,
params2update = {
'Th_universal': 8,
'Th_learned': 9,
'Th_single_ch': detect_threshold, # Whitened Threshold (STD from Absolute traces)
'nt': nt,
'nt0min': snippet_T1,
'dminx': 1,
'nearest_chans': nearest_chans+1,
'max_channel_distance': radius_um,
'do_CAR': False,
'do_correction': False,
'skip_kilosort_preprocessing': sorter_whitening==False,
'whitening_range': nearest_chans+1,
'keep_good_only': True,
'save_extra_vars': True,
'torch_device': 'cuda'
},
estimate_sparsity_params = estimate_sparsity_params,
sorting_analyzer_params = sorting_analyzer_params
)
else:
raise Exception('Setting parameters to run sorter: "{}" is not supported by this pipeline version\nSupported sorters are:\n\t{}\n\t{}'.format(sorterName, 'mountainsort5', 'kilosort4'))
return sorter_info
####################################################################################################################################################################################
# MAIN FUNCTION TO RUN PREPRO:
####################################################################################################################################################################################
# 1) Concatenate & Attach Probe (save Plots)
# 2) Filtering
# 3) Detect Bad Channels
# 4) CommonMedianReference (CMR)
# 5) PowerDensitySpectrum (PSD) (save Plots)
# 6) Peaks Locations | Amplitudes (save plots)
#
# If the number of channels is lower than "nearest_chans" it will NOT run MOTION steps
#
# 7) Motion Estimation (save Plots)
# 8) Motion Interpolation (save Plots: new PeaksLocations)
####################################################################################################################################################################################
def run_prepro(parentRecordingFolder, parentPreproFolder, sessionYear, sessionMonth, sessionDay, electrodeGroup_sessions=None, electrodeGroup_Name=None, session_index=None,
local_radius_chans = (exclude_radius_chans_default, include_radius_chans_default),
noisy_freq = noisy_freq_default,
ms_before = ms_before_default,
ms_after = ms_after_default,
peak_sign = peak_sign_default,
nearest_chans = nearest_chans_default,
peak_detect_threshold = peak_detect_threshold_default,
do_motion = True,
motion_rigid = motion_rigid_default,
motion_options = motion_options_default,
interpolate_options = interpolate_options_default,
localProcess_NWB = False,
rewrite_prepro = True,
return_recording = False
):
if electrodeGroup_sessions is None:
electrodeGroups = SI_tools.getUnique_electrodeGroups(parentRecordingFolder, sessionYear, sessionMonth, sessionDay)
electrodeGroup_sessions = SI_tools.select_electrodeGroup_and_session_info(electrodeGroups, sessionYear, sessionMonth, sessionDay, electrodeGroup_Name, session_index=session_index)
#######################################################################################################################
# Get/Create FOLDER paths to save Figures
#######################################################################################################################
# Create General Session Folder to save Figures
sessionFolder = os.path.join(os.path.abspath(parentPreproFolder), 'exp{}-{:02d}-{:02d}_sorting'.format(sessionYear, sessionMonth, sessionDay))
if not os.path.isdir(sessionFolder):
os.makedirs(sessionFolder)
#############################################################################
# Create Folder to save ElectrodeGroup-Session Preprocessing
elecGroupSessName = '{}-{}'.format(electrodeGroup_sessions['sessID'], electrodeGroup_sessions['electrodeName'])
elecGroupSessFolder = os.path.join(sessionFolder, elecGroupSessName)
if not os.path.isdir(elecGroupSessFolder):
os.mkdir(elecGroupSessFolder)
# Check & clear SI temporal folder (it will be used to save peaks, peaks locations, and NWB files if localProcess_NWB=True)
folder_temp = os.environ.get('SI_PROCESSOR_TEMPDIR')
if folder_temp is None:
SI_tools.set_SItempdir_environ()
folder_temp = os.environ.get('SI_PROCESSOR_TEMPDIR')
SI_tools.clear_SItempdir()
#############################################################################################
# Check if preprocessed Recording exists
if os.path.isfile(os.path.join(elecGroupSessFolder, elecGroupSessName + '_SIrecording.pkl')) and not rewrite_prepro:
print('\nPreprocessed recording was found... \nelectrodeGroup Session : {}\n'.format(elecGroupSessName))
si_recording = si.load_extractor(os.path.join(elecGroupSessFolder, elecGroupSessName + '_SIrecording.pkl'))
step_chan = si_recording.get_annotation('y_contacts_distance_um')
###########################################################
# DETECT PEAKS options:
peaks_options = {
'method': "locally_exclusive",
'peak_sign': peak_sign,
'detect_threshold': peak_detect_threshold, # MAD: Median Amplitude Deviations
'radius_um': step_chan*nearest_chans,
}
###########################################################
# LOCALIZE PEAKS options:
locations_options = {
'ms_before': ms_before,
'ms_after': ms_after,
'location_method': 'monopolar_triangulation', # Paninski Lab
'location_kwargs': {'max_distance_um': step_chan*(nearest_chans + 1), 'optimizer': 'least_square'}
}
else:
#######################################################################################################################
# Copy files if need it & get updated FILE paths to read
#######################################################################################################################
# Check if the NWB files exists in the same Disk as this "code". Otherwise copy NWB-files to the same disk for "speed"
copyFiles = False
if localProcess_NWB:
for f in electrodeGroup_sessions['fileNamePaths']:
filePath, _ = os.path.split(os.path.abspath(f))
if filePath[0:2].lower()!=os.path.abspath(__file__)[0:2].lower():
copyFiles = True
# If need it, copy NWBfiles. Update paths to read files
for f in electrodeGroup_sessions['fileNamePaths']:
fPath, fName = os.path.split(os.path.abspath(f))
if copyFiles:
print('..... copying file {}'.format(fName))
shutil.copy2(os.path.join(fPath, fName), os.path.join(folder_temp, fName))
electrodeGroup_sessions['fileNamePaths_read'].append(os.path.join(folder_temp, fName))
else:
electrodeGroup_sessions['fileNamePaths_read'].append(os.path.join(fPath, fName))
###########################################################
# Save ElectrodeGroupInformation
if not copyFiles:
pickle.dump(electrodeGroup_sessions, open(os.path.join(elecGroupSessFolder, elecGroupSessName + '_electrodeGroupInfo.pkl'), 'wb' ))
print('\nPreprocessing electrodeGroup Session : {}\n'.format(elecGroupSessName))
###################################################
# LOAD session(s)
###################################################
if len(electrodeGroup_sessions['fileNamePaths_read'])>1:
si_recording_List = []
for f in electrodeGroup_sessions['fileNamePaths_read']:
print('Concatenating sessions :\n', f, '\n')
si_recording_List.append(read_nwb(
file_path=f,
electrical_series_path='acquisition/' + electrodeGroup_sessions['electrodePath'],
load_recording=True,
load_sorting=False,
))
si_recording_raw = si.concatenate_recordings(si_recording_List)
else:
f = electrodeGroup_sessions['fileNamePaths_read'][0]
print('loading session : ', f)
si_recording_raw = read_nwb(
file_path=f,
electrical_series_path='acquisition/' + electrodeGroup_sessions['electrodePath'],
load_recording=True,
load_sorting=False,
)
if not si_recording_raw.has_probe():
print('Attaching probe to the recording ......')
probe_from_nwb = SI_tools.constructProbe_2d(si_recordingObj_nwb = si_recording_raw, probeInfo_dict = electrodeGroup_sessions['probeInfo'])
si_recording_rawProbe = si_recording_raw.set_probe(probe_from_nwb)
else:
si_recording_rawProbe = si_recording_raw
del si_recording_raw
###################################################
# Order channels by location
###################################################
print('Sorting channels by depth ......')
chan_ordered_index, chan_orig_index = si.order_channels_by_depth(recording=si_recording_rawProbe, dimensions=("x", "y"))
si_recording_ordered = si_recording_rawProbe.channel_slice(channel_ids=[si_recording_rawProbe.channel_ids[i] for i in chan_ordered_index])
si_recording_ordered.set_property(key='to_raw_index', values = chan_orig_index)
del si_recording_rawProbe
# CONFIRM LOCATIONS ARE IN "um"
if si_recording_ordered.has_probe():
units_probe = numpy.unique(si_recording_ordered._properties['contact_vector']['si_units'])
if len(units_probe)>1:
raise Exception('more than one type of units were found in the location of the contacts within the probe : {}', units_probe)
else:
si_recording_ordered.annotate(location_units=units_probe[0])
# Check if the recording Object has the location_units defined
if 'location_units' in si_recording_ordered.get_annotation_keys():
if si_recording_ordered.get_annotation('location_units') == 'mm':
si_recording_ordered._properties['location'] *=1000
si_recording_ordered._annotations['location_units'] = 'um'
if si_recording_ordered.get_annotation('location_units') != 'mm' and si_recording_ordered.get_annotation('location_units') != 'um':
raise Exception('Units of the Locations in the RecordingObject is not recognized: {}'.format(
si_recording_ordered.get_annotation('location_units')
))
else:
# If not Assume units in 'mm' (default units from NWB format)
si_recording_ordered._properties['location'] *=1000
si_recording_ordered.annotate(location_units='um')
#################################################################################################################
# GET CHANNEL-CONTACT SPACING ("y" coordinate)
contact_locations = si_recording_ordered.get_property('location')
if len(contact_locations)>1:
step_chan = numpy.mean(numpy.absolute(numpy.diff(contact_locations[:, 1])))
else:
step_chan = 50 # Default from SipkeInterface
si_recording_ordered.annotate(y_contacts_distance_um = step_chan)
#################################################################################################################
# ADD CONCATENATION INFORMATION
nSessions = len(electrodeGroup_sessions['fileNames'])
nConcatenations = nSessions-1
if nConcatenations>0:
concatenationSamples = numpy.array(electrodeGroup_sessions['fileSamples'][:-1]).cumsum()
concatenationTimes = concatenationSamples/si_recording_ordered.sampling_frequency
else:
concatenationSamples = numpy.array([])
concatenationTimes = numpy.array([])
si_recording_ordered.annotate(nSessions = nSessions)
si_recording_ordered.annotate(sessionSamples = numpy.array(electrodeGroup_sessions['fileSamples']))
si_recording_ordered.annotate(nConcatenations = nConcatenations)
si_recording_ordered.annotate(concatenationSamples = concatenationSamples)
si_recording_ordered.annotate(concatenationTimes = concatenationTimes)
si_recording_ordered.annotate(savePath = elecGroupSessFolder)
si_recording_ordered.annotate(fileNamePaths = electrodeGroup_sessions['fileNamePaths'])
#################################################################################################################
# PREPROCESSING
#################################################################################################################
#########################################################################################################
# If there is noise at a specefic frequency, remove it with a notch filter
if noisy_freq is not None:
si_recording_denoise = spre.notch_filter(si_recording_ordered, freq=noisy_freq, q=10)
else:
si_recording_denoise = si_recording_ordered
del si_recording_ordered
##########################################################
# High-pass filter
si_recording_filter = spre.bandpass_filter(recording=si_recording_denoise, freq_min=500)
##########################################################
# Median zero-center signal (per channel basis)
si_recording_center_pre = spre.center(recording=si_recording_filter, mode='median')
##########################################################
# Apply Common Median Reference
if si_recording_denoise.get_num_channels()>1:
si_recording_cmr_pre = spre.common_reference(recording=si_recording_center_pre, reference='local', operator='median', local_radius=(local_radius_chans[0] * step_chan, local_radius_chans[1] * step_chan ))
else:
si_recording_cmr_pre = si_recording_center_pre
##########################################################
# Plot session's concatenation
print('\nPlotting {} concatenations from electrodeGroup: {}\n'.format(si_recording_denoise.get_annotation('nConcatenations'), elecGroupSessName))
SI_tools.plot_concatenations(si_recording_dict={'Raw': si_recording_denoise, 'Filter': si_recording_filter, 'CMR': si_recording_cmr_pre},
plot_windows_secs=0.005, sampleChans=True, showPlots=False, savePlots=True, folderPlots=elecGroupSessFolder)
del si_recording_filter, si_recording_cmr_pre
#############################################################################################################
# Detect bad channels
#############################################################################################################
# Use recObj centered (before CMR)
print('\nDetecting Bad Channels ....... \n')
bad_channel_ids, channel_labels = spre.detect_bad_channels(recording=si_recording_center_pre, method="coherence+psd")
# add channel_labels:
# 'coeherence+psd' : good/dead/noise/out
# 'std', 'mad', 'neighborhood_r2' : good/noise
si_recording_center_pre.set_property(key='channel_labels', values=channel_labels)
print('Channel labels : \n\t', channel_labels)
print('\n{} bad channels detected\n\tBad channel Index: {}'.format(len(bad_channel_ids), bad_channel_ids))
##########################################################
# Remove bad channels
si_recording_center_cleanChans = si_recording_center_pre.remove_channels(remove_channel_ids=bad_channel_ids)
del si_recording_center_pre
######################################################################################################
# PLOT THE PROBE & channel labels
print('\nPlotting PROBE from electrodeGroup: {}\n'.format(elecGroupSessName))
fig, ax = plt.subplots(ncols=2, nrows=1, figsize=(12, 8))
ax[0].set_rasterized(True)
plot_probe(probe=si_recording_denoise.get_probe(), ax = ax[0], with_contact_id=False, with_device_index=True)
y_probe = si_recording_denoise.get_channel_locations(axes='y').flatten()
ch_indx = si_recording_denoise.ids_to_indices()
contact_ids = si_recording_denoise.get_channel_ids()
for ch in ch_indx:
ax[1].plot(0, y_probe[ch], marker="s", markeredgecolor=(0, 0, 1, 1), markerfacecolor=(0, 0, 1, 0.5), markeredgewidth = 1, markersize=6)
ax[1].text(-0.5, y_probe[ch], 'ch={}'.format(contact_ids[ch]), horizontalalignment='left', color=(0, 0, 0, 1), fontsize=8)
if channel_labels[ch]=='good':
colorT = (0, 0, 1, 1)
fweight = 'normal'
else:
colorT = (1, 0, 0, 1)
fweight = 'bold'
ax[1].text(0.5, y_probe[ch], channel_labels[ch], horizontalalignment='right', color=colorT, fontsize=8, fontweight=fweight)
ax[1].set_xlim(-1, 1)
ax[1].set_rasterized(True)
fig.savefig(os.path.join(elecGroupSessFolder, elecGroupSessName + '_probe.eps'), dpi='figure', format='eps')
plt.close(fig=fig)
del bad_channel_ids, channel_labels, si_recording_denoise
#####################################################################
# PLOT Power Spectrum Density
print('\nPlotting PowerDensitySpectrums from electrodeGroup: {}\n'.format(elecGroupSessName))
SI_tools.plotPSD_randomChunks(si_recording_center_cleanChans, compare_CMR=True, plot_by_channel=True, chan_radius=local_radius_chans, showPlots=False, savePlots=True, folderPlots=elecGroupSessFolder)
#############################################################################################################
# Reference after removing bad channels
#############################################################################################################
if si_recording_center_cleanChans.get_num_channels()>1:
si_recording = spre.common_reference(recording=si_recording_center_cleanChans, reference='local', operator='median', local_radius=(local_radius_chans[0] * step_chan, local_radius_chans[1] * step_chan ))
else:
si_recording = si_recording_center_cleanChans
###################################################
# Add some annotations:
si_recording.annotate(is_filtered=True)
si_recording.annotate(is_centered=True)
si_recording.annotate(centered_mode = 'median')
si_recording.annotate(is_referenced=True)
si_recording.annotate(reference = 'local')
si_recording.annotate(reference_mode = 'median')
del si_recording_center_cleanChans
####################################################################################################################################
# If NWB files were not move from original location: Save Recording Object (LAZY without traces)
if not copyFiles:
si_recording.dump_to_pickle(os.path.join(elecGroupSessFolder, elecGroupSessName + '_SIrecording'))
########################################################################################################################
# PEAKS LOCATIONS AS A FUNCTION OF TIME
########################################################################################################################
###########################################################
# DETECT PEAKS options:
peaks_options = {
'method': "locally_exclusive",
'peak_sign': peak_sign,
'detect_threshold': peak_detect_threshold, # MAD: Median Amplitude Deviations
'radius_um': step_chan*nearest_chans,
}
###########################################################
# LOCALIZE PEAKS options:
locations_options = {
'ms_before': ms_before,
'ms_after': ms_after,
'location_method': 'monopolar_triangulation', # Paninski Lab
'location_kwargs': {'max_distance_um': step_chan*(nearest_chans + 1), 'optimizer': 'least_square'}
}
###########################################################
# PLOT PEAKS LOCATIONS AS A FUNCTION OF TIME
print('\nPlotting Peaks Locations from electrodeGroup: {}\n'.format(elecGroupSessName))
SI_tools.plot_peakLocations(
si_recording = si_recording,
folderPeaks = folder_temp,
extra_recording_label = '',
peaks_options = peaks_options,
locations_options = locations_options,
rewrite = False,
locationsSubSampled = True,
savePlots = True,
showPlots = False,
folderPlots = elecGroupSessFolder
)
########################################################################################################################
# NOTE:
# IF CHANNEL COUNT IS LOWER THAN "nearest_chans" IT WILL NOT RUN MOTION CORRECTION
if si_recording.get_num_channels()>=nearest_chans and do_motion:
########################################################################################################################
# MOTION ESTIMATION & INTERPOLATION
########################################################################################################################
#####################################################################
# Get peaks labels:
if "by_channel" in peaks_options['method']:
peaks_prefix = 'byCh'
elif "locally_exclusive" in peaks_options['method']:
peaks_prefix = 'loc'
else:
raise Exception('Peaks Detection method "{}" not recognized\nAvailable options: {}'.format(peaks_options['method'], ["by_channel", "locally_exclusive", "locally_exclusive_cl",
"by_channel_torch", "locally_exclusive_torch", "matched_filtering"]))
peaks_label = '{}{}{}'.format(peaks_prefix, peaks_options['peak_sign'][0].upper(), peaks_options['detect_threshold']).replace('.', '_')
#####################################################################
# Get location labels:
if locations_options['location_method']=='center_of_mass':
loc_label = 'mass'
elif locations_options['location_method']=='monopolar_triangulation':
loc_label = 'mono'
elif locations_options['location_method']=='grid_convolution':
loc_label = 'grid'
else:
raise Exception('Peaks Location method "{}" not recognized\nAvailable options: {}'.format(peaks_options['method'], ['center_of_mass', 'monopolar_triangulation', 'grid_convolution']))
peaksLoc_label = peaks_label + '_' + loc_label
#####################################################################
# Get motion labels:
if motion_rigid:
rigid_label = 'rigid'
win_step_um = step_chan/2
else:
rigid_label = 'noRigid'
win_step_um = step_chan
if motion_options['method']=='decentralized':
motion_label = 'deCENTRAL' + rigid_label
elif motion_options['method']=='iterative_template':
motion_label = 'iterTEMP' + rigid_label
elif motion_options['method']=='dredge_ap':
motion_label = 'dredgeAP' + rigid_label
elif motion_options['method']=='dredge_lfp':
motion_label = 'dredgeLFP' + rigid_label
else:
raise Exception('Motion Estimation method "{}" not recognized\nAvailable options: {}'.format(motion_options['method'], ['decentralized', 'iterative_template', 'dredge_ap', 'dredge_lfp']))
#####################################################################
# Get interpolation labels:
if interpolate_options['method']=='idw':
interpolation_label = 'idw'
elif interpolate_options['method']=='nearest':
interpolation_label = 'near'
elif interpolate_options['method']=='kriging':
interpolation_label = 'krig'
else:
raise Exception('Interpolation method "{}" not recognized\nAvailable options: {}'.format(interpolate_options['method'], ['kriging', 'idw', 'nearest']))
if interpolate_options['border_mode']=="remove_channels":
border_label = 'Rmv'
elif interpolate_options['border_mode']== "force_extrapolate":
border_label = 'Extrap'
elif interpolate_options['border_mode']=="force_zeros":
border_label = 'Zeros'
else:
raise Exception('BorderMode method "{}" not recognized\nAvailable options: {}'.format(interpolate_options['border_mode'], ['remove_channels', 'force_extrapolate', 'force_zeros']))
motion_interpolation_label = motion_label + '_' + interpolation_label + border_label
si_recording_sufix = '_' + peaksLoc_label + '_' + motion_interpolation_label
if os.path.isfile(os.path.join(elecGroupSessFolder, elecGroupSessName + si_recording_sufix + '_SIrecording' + '.pkl')) and not rewrite_prepro:
print('\nMotion corrected recording was found... \nelectrodeGroup Session : {}\nPeaks {}: sign = {}, detectionTH = {}\nLocation Method: {}\nMotion Interpolation label : {}\n'.format(elecGroupSessName,
peaks_options['method'], peaks_options['peak_sign'], peaks_options['detect_threshold'], locations_options['location_method'], motion_interpolation_label))
si_recording_motion = si.load_extractor(os.path.join(elecGroupSessFolder, elecGroupSessName + si_recording_sufix + '_SIrecording' + '.pkl'))
else:
##################################################################
# Load PEAKS
if not os.path.exists(os.path.join(folder_temp, 'peaks_' + peaks_label + '.npy')):
print('\nGetting Peaks from electrodeGroup: {}\n'.format(elecGroupSessName))
noise_levels = si.get_noise_levels(si_recording, return_scaled=False)
peaks = detect_peaks(recording=si_recording,
noise_levels = noise_levels,
method=peaks_options['method'],
gather_mode = 'memory', # gather_mode= 'npy', # 'npy'
folder = None, # folder = eg_dirs['motion'],
names = None, # names = ['peaks'],
peak_sign=peaks_options['peak_sign'],
detect_threshold=peaks_options['detect_threshold'],
radius_um=peaks_options['radius_um'],
**job_kwargs)
numpy.save(os.path.join(folder_temp, 'peaks_' + peaks_label + '.npy'), peaks)
del peaks, noise_levels
peaks = numpy.load(os.path.join(folder_temp, 'peaks_' + peaks_label + '.npy'))
##################################################################
# Load all PEAKS LOCATIONS
if not os.path.exists(os.path.join(folder_temp, 'peaks_' + peaks_label + '_locations_' + loc_label + '.npy')):
print('\nGetting Peaks Locations from electrodeGroup: {}\n'.format(elecGroupSessName))
peaks_locations = localize_peaks(
recording=si_recording,
peaks=peaks,
ms_before= locations_options['ms_before'],
ms_after= locations_options['ms_after'],
radius_um = peaks_options['radius_um'],
method= locations_options['location_method'],
**locations_options['location_kwargs'],
**job_kwargs
)
numpy.save(os.path.join(folder_temp, 'peaks_' + peaks_label + '_locations_' + loc_label + '.npy'), peaks_locations)
del peaks_locations
peaks_locations = numpy.load(os.path.join(folder_temp, 'peaks_' + peaks_label + '_locations_' + loc_label + '.npy'))
###############################################################################
# MOTION ESTIMATION
print('\nEstimating MOTION from electrodeGroup: {}\n'.format(elecGroupSessName))
motion, extra_check = estimate_motion(
# Parameters common to all motion Methods
recording=si_recording,
peaks=peaks,
peak_locations=peaks_locations,
direction="y", # **histogram section**
rigid=motion_rigid, # **non-rigid section** # default False
win_shape="gaussian", # default "gaussian"
win_step_um = win_step_um, # default 50.0
win_scale_um = win_step_um*3.0, # default 150.0
win_margin_um = None, # default -win_scale_um/2
method = motion_options['method'], # **method options**
extra_outputs=True, # **extra options**
progress_bar=True,
verbose=False,
# Parameters common to all motion Methods but defined as **method_kwargs
bin_um = step_chan/4, # default 10.0
bin_s = 10.0, # default 10.0 # bin_s=10.0, # default 10.0
# Specific Method arguments
**motion_options['method_kwargs'],
)
contact_locations = si_recording.get_property('location')
minY_loc = -50
maxY_loc = max(contact_locations[:, 1])+50
SI_tools.plot_motion_outputs(
peaks = peaks,
peaks_locations = peaks_locations,
sampling_frequency = si_recording.sampling_frequency,
motion = motion,
extra_check = extra_check,
peaks_label = peaks_label,
loc_label = loc_label,
motion_label = motion_label,
minY_loc = minY_loc,
maxY_loc = maxY_loc,
folderPlots = elecGroupSessFolder,
prefixRec = elecGroupSessName,
showPlots = False,
savePlots = True,
concatenationTimes = si_recording.get_annotation('concatenationTimes'),
verbose = True
)
####################################################################################################################################
# Interpolate motion
####################################################################################################################################
##################################################################
print('motion interpolaton : ...' + peaks_label + '-' + loc_label + '-' + motion_interpolation_label + '....')
si_recording_motion = interpolate_motion(
recording = si_recording,
motion = motion,
border_mode = interpolate_options['border_mode'],
spatial_interpolation_method = interpolate_options['method'],
sigma_um = step_chan, # Used in the "kriging" formula
p=1, # Used in the "kriging" formula
num_closest=3, # Number of closest channels used by "idw" method for interpolation
)
si_recording_motion.annotate(motion_corrected = True)
si_recording_motion.annotate(motion_method_label = si_recording_sufix[1:])
###########################################################
# PLOT NEW PEAKS LOCATIONS AS A FUNCTION OF TIME
print('\nPlotting Motion Corrected Peaks Locations from electrodeGroup: {}\n'.format(elecGroupSessName))
SI_tools.plot_peakLocations(
si_recording = si_recording_motion,
folderPeaks = folder_temp,