-
Notifications
You must be signed in to change notification settings - Fork 1
/
running_script.R
1557 lines (1415 loc) · 50.9 KB
/
running_script.R
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
#### 2.1 R packages & versioning ####
## Specify the packages you'll use in the script
packages <- c("tidyverse",
"zoo",
"gridExtra",
"R.matlab",
"cowplot",
"easystats",
"circular",
"splines",
"MESS", ## area under curve
"zoo" ## rolling means
)
## Now for each package listed, first check to see if the package is already
## installed. If it is installed, it's simply loaded. If not, it's downloaded
## from CRAN and then installed and loaded.
package.check <- lapply(packages,
FUN = function(x) {
if (!require(x, character.only = TRUE)) {
install.packages(x, dependencies = TRUE)
library(x, character.only = TRUE)
}
}
)
#### 2.2 %not_in% ####
`%not_in%` <- Negate(`%in%`)
#### 5 Data wrangling ####
#### 5.1.1 Identify files to import ####
## List all files of each file type
csv_files <-
list.files("./data", pattern = ".csv",
full.names = TRUE)
mat_files <-
list.files("./data", pattern = ".mat",
full.names = TRUE)
## Generate metadata tibbles for each file type
csv_file_info <-
tibble(
csv_files = csv_files,
## Extract the basename by removing the file extension
basename = basename(csv_files) %>% str_remove(".csv"),
## NOTE: PLEASE ADJUST THE FOLLOWING LINE TO BE ABLE TO EXCTRACT OUT THE
## DATE BASED ON YOUR NAMING CONVENTION
basedate = basename(csv_files) %>% str_sub(start = 1, end = 12)
)
mat_file_info <-
tibble(
mat_files = mat_files,
## Extract the basename by removing the file extension
basename = basename(mat_files) %>% str_remove(".mat"),
## NOTE: AGAIN, PLEASE ADJUST THE FOLLOWING LINE TO BE ABLE TO EXCTRACT OUT
## THE DATE BASED ON YOUR NAMING CONVENTION
basedate = basename(mat_files) %>% str_sub(start = 1, end = 12)
)
## Matchmake between .MAT data and .CSV log files
csv_mat_filejoin <-
inner_join(csv_file_info, mat_file_info, by = "basename") %>%
## OPTIONAL STEP: remove any rows where either the .MAT or .CSV is missing
drop_na()
## Store a vector of basenames in the environment. This will become useful later
base_names <- csv_mat_filejoin$basename
#### 5.1.2 Data import and preliminary labeling ####
## Set up empty vectors that will collect sets of replicates that we will be
## splitting up
metadata_sets <- NULL
meta_splits <- NULL
data_splits <- NULL
gc()
starttime <- Sys.time() ## Optional, will help you assess run time
for (i in 1:nrow(csv_mat_filejoin)) {
## Which file # are we working on?
print(i)
## Set up temporary objects in which to eventually write data
csv_data_sets <- NULL
mat_data_sets <- NULL
joined_data_sets <- NULL
## Import the matlab file. This may take some time
mat_import <-
R.matlab::readMat(csv_mat_filejoin[i,"mat_files"])
## Read in the corresponding csv log file
csv_data_sets[[i]] <-
read_csv(as.character(csv_mat_filejoin[i,"csv_files"]),
show_col_types = FALSE) %>%
## Rename columns for convenience
rename(
Spatial_Frequency = `Spatial Frequency`,
Temporal_Frequency = `Temporal Frequency`,
Cycles_Per_Pixel = `Cycles Per Pixel`
)
## Determine if spatial frequencies need to be transformed
sfs <- csv_data_sets[[i]]$Spatial_Frequency %>% unique() %>% sort()
cpps <- c(0.000668, 0.001336, 0.002670, 0.005300, 0.010600, 0.021200)
if (all(sfs == cpps)) {
## If true, convert to cpd using the following mapping:
csv_data_sets[[i]]$Spatial_Frequency[csv_data_sets[[i]]$Spatial_Frequency == 0.000668] <-
2^-6
csv_data_sets[[i]]$Spatial_Frequency[csv_data_sets[[i]]$Spatial_Frequency == 0.001336] <-
2^-5
csv_data_sets[[i]]$Spatial_Frequency[csv_data_sets[[i]]$Spatial_Frequency == 0.00267] <-
2^-4
csv_data_sets[[i]]$Spatial_Frequency[csv_data_sets[[i]]$Spatial_Frequency == 0.0053] <-
2^-3
csv_data_sets[[i]]$Spatial_Frequency[csv_data_sets[[i]]$Spatial_Frequency == 0.0106] <-
2^-2
csv_data_sets[[i]]$Spatial_Frequency[csv_data_sets[[i]]$Spatial_Frequency == 0.0212] <-
2^-1
}
## The log file does not have time = 0, so set up a separate tibble to
## add this info in later. Some of the metadata will just be filler for now.
initial <- tibble(
Trial = "initialization",
Spatial_Frequency = csv_data_sets[[i]]$Spatial_Frequency[1],
Cycles_Per_Pixel = csv_data_sets[[i]]$Cycles_Per_Pixel[1],
Temporal_Frequency = csv_data_sets[[i]]$Temporal_Frequency[1],
Direction = csv_data_sets[[i]]$Direction[1],
Time = 0.000
)
## Find photodiode
## It is almost always in channel 2, but we should be sure to check before
## extracting automatically
photod_default_channel <-
mat_import[[stringr::str_which(names(mat_import), "Ch2")[1]]]
if (!photod_default_channel[1][[1]][1] == "waveform") {
warning("File ", i,": Photodiode channel identity uncertain")
}
## Find spikes
## Similarly, spikes are almost always in channel 5, but we should check
spikes_default_channel <-
mat_import[[stringr::str_which(names(mat_import), "Ch5")[1]]]
if("codes" %not_in% attributes(spikes_default_channel)$dimnames[[1]]) {
warning("File ", i,": Sorted spikes channel identity uncertain")
}
## If that worked, see if we can automatically determine the "times" and
## "codes" slot numbers
times_location <-
which(attributes(spikes_default_channel)$dimnames[[1]] == "times")
codes_location <-
which(attributes(spikes_default_channel)$dimnames[[1]] == "codes")
## Find matlab's stimulus change log
stim_change_channel <-
mat_import[[stringr::str_which(names(mat_import), "Ch3")[1]]]
## Each sweep should be 5 secs. We'll check that the median is 5
## If this results in an error, then the channel identity could be wrong, or
## there may have been an issue with sweep duration during the recording
## process
if(!median(round(diff(stim_change_channel[[5]][,1]),0)) == 5) {
warning("File ", i,": stim change channel identity uncertain")
}
## Determine when the onset of motion occurred according to matlab
first_moving_mat <-
stim_change_channel[[5]][,1][1]
## Find the first "moving" phase in the log file
first_moving_csv <-
csv_data_sets[[i]] %>%
filter(Trial == "moving") %>%
select(Time) %>%
slice(1) %>%
as.numeric()
## Find the first "blank" phase in the log file
first_blank <-
csv_data_sets[[i]] %>%
filter(Trial == "blank") %>%
select(Time) %>%
slice(1) %>%
as.numeric()
## Compute the difference between these two
first_mvbl_diff <- first_moving_csv - first_blank
## Check to see if the final row of the metadata is "moving" and truncate
## if not
## This can effectively be done by truncating after the final "moving" phase
max_moving_sweep <-
max(which(csv_data_sets[[i]]$Trial == "moving"))
first_csv_tmp <-
bind_rows(initial, csv_data_sets[[i]]) %>%
## Add 1 to max moving sweep since we tacked on "initial" in previous step
## Then slice to restrict any extraneous partial sweeps
slice_head(n = (max_moving_sweep + 1)) %>%
## Add the first event time to "Time" and subtract first_mvbl_diff (~2 secs)
## What this does is shift the log csv's time stamping to match the matlab
## file's stim change channel's time stamping
mutate(Time = Time + first_moving_mat - first_mvbl_diff - first_blank) %>%
## Make character version of Time for joining later
## This will be crucial for _join functions
mutate(Time_char = as.character(round(Time,3)))
## Duplicate the initialization for ease of setting T0
inception <-
initial %>%
mutate(Time_char = as.character(round(Time,3)))
inception$Trial[1] <- "inception"
## Bind the initialization rows
first_csv <-
bind_rows(inception, first_csv_tmp)
## Compute stimulus end times
first_csv$Stim_end <- c(first_csv$Time[-1], max(first_csv$Time) + 3)
## Get final time
final_time <- first_csv$Stim_end[nrow(first_csv)]
## Extract photodiode data
## First generate a time sequence to match to the photodiode trace
Time_vec <- seq(
from = 0.0000,
by = 1 / 25000,
length.out = length(photod_default_channel[9][[1]][, 1])
)
## The key thing is to get a character version of time from this
Time_char_vec <- as.character(round(Time_vec, 3))
## Grab the photodiode data
photod_full <-
tibble(Photod =
photod_default_channel[9][[1]][, 1])
## Add numeric time
photod_full$Time <-
seq(
from = 0.0000,
by = 1 / 25000,
length.out = nrow(photod_full)
)
options(scipen = 999)
photod_full <-
photod_full %>%
## Add the character time
add_column(Time_char = Time_char_vec) %>%
## Use the charcter time to define a group
group_by(Time_char) %>%
## Then average the photodiode within
summarise(Photod = mean(Photod)) %>%
ungroup() %>%
mutate(Time = round(as.numeric(Time_char), 3)) %>%
arrange(Time) %>%
filter(Time <= final_time)
## Extract all spike data
all_spike_dat <-
tibble(
Time =
spikes_default_channel[times_location][[1]][, 1],
code =
spikes_default_channel[codes_location][[1]][, 1]) %>%
## Characterize time, for purposes of joining later
mutate(Time_char = as.character(round(Time, 3)))
## How many distinct neurons are there?
cell_ids <- sort(unique(all_spike_dat$code))
n_cells <- 1:length(cell_ids)
if(length(n_cells) > 1) { ## if there's more than one distinct neuron
all_spike_dat_tmp <-
all_spike_dat %>%
## Group by identity of spiking neuron
group_by(code) %>%
## Split into separate dfs, one per neuron
group_split()
## Additional cells are labeled as "Spike_n"
all_cells <- NULL
for (j in n_cells) {
#print(j)
new_name = paste0("Spikes_", cell_ids[j])
all_cells[[j]] <-
all_spike_dat_tmp[[j]]
## Consolidate to 3 decimal places
all_cells[[j]] <-
all_cells[[j]] %>%
group_by(Time_char) %>%
summarise(code = mean(code)) %>%
mutate(code = ceiling(code)) %>%
ungroup() %>%
mutate(Time = round(as.numeric(Time_char), 3)) %>%
arrange(Time) %>%
filter(Time <= final_time)
names(all_cells[[j]])[match("code", names(all_cells[[j]]))] <-
new_name
## Replace "j" with 1 to indicate presence/absence of spike rather than
## cell identity
all_cells[[j]][new_name] <- 1
## If the identity is 1, replace "Spikes_1" with just "Spikes"
if (new_name == "Spikes_1") {
names(all_cells[[j]])[match(new_name, names(all_cells[[j]]))] <-
"Spikes"
}
}
## Consolidate
all_spike_dat <-
all_cells %>%
## Tack on additional spike columns
reduce(full_join, by = "Time_char") %>%
arrange(Time_char) %>%
## Remove time.n columns but
## Do not remove Time_char
select(-contains("Time.")) %>%
## Regenerate numeric time
mutate(
Time = as.numeric(Time_char)
) %>%
select(Time, Time_char, Spikes, everything()) %>%
filter(Time <= final_time)
} else { ## If there's only 1 neuron
all_spike_dat <-
all_spike_dat %>%
group_by(Time_char) %>%
summarise(code = mean(code)) %>%
mutate(code = ceiling(code)) %>%
ungroup() %>%
rename(Spikes = code) %>%
mutate(Time = round(as.numeric(Time_char), 3)) %>%
arrange(Time) %>%
filter(Time <= final_time) %>%
select(Time, Time_char, everything())
}
options(scipen = 999)
mat_data_sets[[i]] <-
## Generate a time sequence from 0 to final_time
tibble(
Time = seq(from = 0, to = final_time, by = 0.001)
) %>%
## Character-ize it
mutate(Time_char = as.character(round(Time, 5))) %>%
## Join in the photodiode data
left_join(photod_full, by = "Time_char") %>%
select(-Time.y) %>%
rename(Time = Time.x) %>%
arrange(Time) %>%
## Join in the spike data
left_join(all_spike_dat, by = "Time_char") %>%
select(-Time.y) %>%
rename(Time = Time.x) %>%
arrange(Time) %>%
filter(Time <= final_time) %>%
## Replace NAs with 0 in the Spike columns only
mutate(
across(starts_with("Spikes"), ~replace_na(.x, 0))
)
## Merge the matlab data with the metadata
joined_one_full <-
mat_data_sets[[i]] %>%
## Join by the character version of time NOT the numerical!!
full_join(first_csv, by = "Time_char") %>%
## Rename columns for clarity of reference
rename(Time_mat = Time.x,
Time_csv = Time.y) %>%
## Convert character time to numeric time
mutate(Time = round(as.numeric(Time_char), 3)) %>%
## Carry metadata forward
mutate(
Trial = zoo::na.locf(Trial, type = "locf"),
Spatial_Frequency = zoo::na.locf(Spatial_Frequency, type = "locf"),
Cycles_Per_Pixel = zoo::na.locf(Cycles_Per_Pixel, type = "locf"),
Temporal_Frequency = zoo::na.locf(Temporal_Frequency, type = "locf"),
Direction = zoo::na.locf(Direction, type = "locf"),
Time_csv = zoo::na.locf(Time_csv, type = "locf"),
Stim_end = zoo::na.locf(Stim_end, type = "locf")
) %>%
## Calculate velocity
mutate(
Speed = round(Temporal_Frequency/Spatial_Frequency, 0),
Log2_Speed = log2(Speed)
)
## Add info to metadata
metadata_one_full <-
first_csv %>%
mutate(
Speed = round(Temporal_Frequency/Spatial_Frequency, 0),
Log2_Speed = log2(Speed),
Stim_end_diff = c(0, diff(Stim_end))
)
## Some quality control checks
## What are the stim time differences?
stimtime_diffs <- round(metadata_one_full$Stim_end_diff)[-c(1:2)]
## How many total reps were recorded?
stimtime_reps <- length(stimtime_diffs)/3
## What do we expect the overall structure to look like?
stimtime_expectation <- rep(c(1,1,3), stimtime_reps)
## Does reality match our expectations?
if (!all(stimtime_diffs == stimtime_expectation)) {
## If you get this, investigate the file further and determine what went
## wrong
print("stimtime issue; investigate")
}
## Sometimes the final sweep gets carried for an indefinite amount of time
## before the investigator manually shuts things off. The following block
## truncates accordingly
mark_for_removal <-
which(round(metadata_one_full$Stim_end_diff) %not_in% c(1, 3))
if (any(mark_for_removal == 1 | mark_for_removal == 2)) {
mark_for_removal <- mark_for_removal[mark_for_removal > 2]
}
if (length(mark_for_removal) > 0) {
metadata_sets[[i]] <-
metadata_one_full %>%
filter(Time < metadata_one_full[mark_for_removal[1],]$Time)
joined_data_sets[[i]] <-
joined_one_full %>%
filter(Time_mat < metadata_one_full[mark_for_removal[1],]$Time)
} else {
metadata_sets[[i]] <-
metadata_one_full
joined_data_sets[[i]] <-
joined_one_full
}
## Organize the metadata for export in the R environment
meta_splits[[i]] <-
metadata_sets[[i]] %>%
## Get rid of the non-sweep info
filter(!Trial == "inception") %>%
filter(!Trial == "initialization") %>%
## Group by trial
#group_by(Spatial_Frequency, Temporal_Frequency, Direction) %>%
## Split by trial group
group_split(Spatial_Frequency, Temporal_Frequency, Direction)
data_splits[[i]] <-
joined_data_sets[[i]] %>%
## Get rid of the non-sweep info
filter(!Trial == "inception") %>%
filter(!Trial == "initialization") %>%
## Group by trial
group_by(Spatial_Frequency, Temporal_Frequency, Direction) %>%
## Split by trial group
group_split()
## Do some cleanup so large objects don't linger in memory
rm(
first_csv, inception, initial, mat_import, first_csv_tmp,
photod_default_channel, spikes_default_channel, photod_full,
all_spike_dat, all_spike_dat_tmp, all_cells,
metadata_one_full, joined_one_full, joined_data_sets,
csv_data_sets, mat_data_sets
)
message("File ", i, ": ", csv_mat_filejoin[i,"basename"], " imported")
gc()
}
endtime <- Sys.time()
endtime - starttime ## Total elapsed time
## Tidy up how R has been using RAM by running garbage collection
gc()
## Name each data set according to the basename of the file
names(metadata_sets) <- csv_mat_filejoin$basename #base_names
names(meta_splits) <- csv_mat_filejoin$basename #base_names
names(data_splits) <- csv_mat_filejoin$basename #base_names
## Old method: Extract matlab spike and photodiode data
# #mat_data_sets[[i]] <-
# spikedat <-
# data.frame(
# Time =
# seq(
# from = 0, by = 0.001,
# length.out = 1 + length(t(mat_import$spikes))
# ), #t(mat_import$tt),
# Spikes = c(0, t(mat_import$spikes)),
# Photod = c(0, mat_import$photodiode)
# ) %>%
# as_tibble() %>%
# mutate(Time_char = as.character(Time)) %>%
# filter(Time <= final_time)
## Get organized lists of stimuli that were used
## This will ultimately be used for arranging data by stimuli in a sensible
## order
metadata_combos <- NULL
for (i in 1:length(metadata_sets)) {
metadata_combos[[i]] <-
metadata_sets[[i]] %>%
## Get unique stimulus parameters
distinct(Spatial_Frequency, Temporal_Frequency, Speed, Direction) %>%
arrange(Direction) %>%
## Sort by SF (smallest to largest)
arrange(desc(Spatial_Frequency)) %>%
mutate(
## Set a "plot_order" which provides a running order of stimuli
plot_order = 1:length(meta_splits[[i]]),
name = paste(Direction, "Deg,", Speed, "Deg/s")
)
}
names(metadata_combos) <- csv_mat_filejoin$basename #base_names
#### 5.2 Organizing replicates (required) and binning (optional) ####
#### __SET BIN SIZE HERE ####
## Set bin size here
## Units are in ms (e.g. 10 = 10ms)
bin_size = 10 ## 10 or 100 or 1 (1 = "unbinned")
slice_size = NULL
slicemin = NULL
slicemax = NULL
condition = NULL
if (bin_size == 10){
slice_size <- 501
slicemin <- 202
slicemax <- 498
condition <- "_binsize10"
} else if (bin_size == 100){
slice_size <- 51
slicemin <- 21
slicemax <- 49
condition <- "_binsize100"
} else if (bin_size == 1){
slice_size <- NULL
slicemin <- NULL
slicemax <- NULL
condition <- "_unbinned"
} else {
stop("bin_size is non-standard")
}
all_replicate_data_reorganized <-
vector(mode = "list", length = length(meta_splits))
name_sets <-
vector(mode = "list", length = length(meta_splits))
gc()
starttime <- Sys.time()
for (i in 1:length(meta_splits)){
## i = file number
print(i)
## We'll need to collect data at a per-stimulus level and on a per-replicate
## level within the per-stimulus level
## "j" will be used to designate a unique stimulus
## We'll first create an empty object in which to collect stimulus-specific
## data
replicate_data_reorganized <- NULL
## For each of j unique stimuli...
for (j in 1:length(meta_splits[[i]])) { # j = {direction,speed}
## Isolate the j-th data
d <- data_splits[[i]][[j]]
## And the j-th log data
m <- meta_splits[[i]][[j]] %>%
group_by(Trial) %>%
## Label separate replicates
mutate(Replicate = row_number())
## Extract a stimulus label to a name_set that will be used later
name_sets[[i]][[j]] <-
paste(m$Direction[1], "Deg,", m$Speed[1], "Deg/s")
## Set up a temporary object to deal with per-replicate data
replicates_ordered <- NULL
## "k" will be used to designate replicate number
for (k in 1:max(m$Replicate)){
tmp <-
m %>%
filter(Replicate == k)
## If you have a complete replicate (i.e., blank, stationary, moving)
if (nrow(tmp) == 3 ) {
## Grab the specific data
doot <-
d %>%
filter(Time >= min(tmp$Time)) %>%
filter(Time <= max(tmp$Stim_end))
## Add bin information
doot$bin <-
rep(1:ceiling(nrow(doot)/bin_size), each = bin_size)[1:nrow(doot)]
if (bin_size == 1) {
## IF YOU ARE NOT BINNING, RUN THIS:
replicates_ordered[[k]] <-
doot %>%
mutate(
## Construct a standardized time within the sweep
Time_stand = Time_mat - min(Time_mat),
## When does the sweep begin
Time_begin = min(Time_mat),
## When does the sweep end
Time_end = max(Time_mat),
## Delineate the end of the blank phase
Blank_end = tmp$Stim_end[1] - min(Time_mat),
## Delineate the end of the stationary phase
Static_end = tmp$Stim_end[2] - min(Time_mat),
## Label the replicate number
Replicate = k
) %>%
## Bring stim info to first few columns
select(Speed, Spatial_Frequency, Temporal_Frequency, Direction,
everything()) %>%
## Just in case there some hang over
filter(Time_stand >= 0)
} else { ## IF YOU ARE BINNING, RUN THIS:
## First grab time and meta info
time_and_meta <-
doot %>%
## WITHIN EACH BIN:
group_by(bin) %>%
summarise(
## Label the trial
Trial = first(Trial),
## Midpoint of bin
Time_bin_mid = mean(Time_mat),
## Bin beginning
Time_bin_begin = min(Time_mat),
## Bin end
Time_bin_end = max(Time_mat),
## Spike rate = sum of spikes divided by elapsed time
Spike_rate = sum(Spikes) / (max(Time_mat) - min(Time_mat)),
Photod_mean = mean(Photod)
)
## Now deal with Spike_N columns
hold_spike_n <-
doot %>%
select(starts_with("Spikes_")) %>%
add_column(bin = doot$bin) %>%
add_column(Time_mat = doot$Time_mat) %>%
group_by(bin) %>%
summarise(across(starts_with("Spikes_"),
~ sum(.x) / (max(Time_mat) - min(Time_mat))))
## Put them together
replicates_ordered[[k]] <-
time_and_meta %>%
left_join(hold_spike_n, by = "bin") %>%
mutate(
## Add in metadata (following same definitions above)
Time_stand = Time_bin_mid - min(Time_bin_mid),
Blank_end = tmp$Stim_end[1] - min(Time_bin_mid),
Static_end = tmp$Stim_end[2] - min(Time_bin_mid),
Spatial_Frequency = m$Spatial_Frequency[1],
Temporal_Frequency = m$Temporal_Frequency[1],
Speed = m$Speed[1],
Direction = m$Direction[1],
Replicate = k
) %>%
## Bring stim info to first few columns
select(Speed, Spatial_Frequency, Temporal_Frequency, Direction,
everything()) %>%
## Just in case there some hang over
filter(Time_stand >= 0) %>%
filter(bin < slice_size + 1)
rm(time_and_meta, hold_spike_n)
}
}
}
## Now insert it within the collector of per-stimulus data
replicate_data_reorganized[[j]] <-
replicates_ordered %>%
bind_rows()
## Now insert it within the overall data collector
all_replicate_data_reorganized[[i]][[j]] <-
replicate_data_reorganized[[j]]
## Toss out temporary objects and clean up
rm(replicates_ordered, d, m, tmp)
gc()
}
}
endtime <- Sys.time()
endtime - starttime
gc()
for (i in 1:length(all_replicate_data_reorganized)) {
for (j in 1:length(all_replicate_data_reorganized[[i]])) {
names(all_replicate_data_reorganized[[i]])[[j]] <- name_sets[[i]][[j]]
}
}
names(all_replicate_data_reorganized) <- csv_mat_filejoin$basename #base_names
## check to see if all averaged replicate sets are in the same order with
## respect to direction and speed
names_grid <- matrix(
ncol = length(all_replicate_data_reorganized),
nrow = length(all_replicate_data_reorganized)
)
for (i in 1:length(all_replicate_data_reorganized)) {
for (j in 1:length(all_replicate_data_reorganized)) {
names_grid[i,j] <-
identical(names(all_replicate_data_reorganized[[i]]),
names(all_replicate_data_reorganized[[j]])
)
}
}
#### 5.3 Data export ####
## Export a csv for each session with all data organized
## Declare export destination
export_path <- "./data/"
## The "condition" will be appended to the file name.
## Export each tibble within all_replicate_data_reorganized
for (i in 1:length(all_replicate_data_reorganized)) {
print(i)
dat <-
all_replicate_data_reorganized[[i]] %>%
bind_rows()
write_csv(
dat,
file =
paste0(
export_path,
names(all_replicate_data_reorganized)[i],
condition,
".csv"
)
)
rm(dat)
}
gc()
#### 6 Raster and mean spike rate plots ####
#### 6.1 Data sets ####
## File paths and basenames of _unbinned.csv files
unbinned_filelist <-
list.files("./data/", pattern = "_unbinned.csv",
full.names = TRUE)
unbinned_basenames <-
unbinned_filelist %>%
str_remove("./data/") %>%
str_remove("_unbinned.csv")
## File paths and basenames of _binsize10.csv files
bin10_filelist <-
list.files("./data/", pattern = "_binsize10.csv",
full.names = TRUE)
bin10_basenames <-
bin10_filelist %>%
str_remove("./data/") %>%
str_remove("_binsize10.csv")
## File paths and basenames of _binsize100.csv files
bin100_filelist <-
list.files("./data/", pattern = "_binsize100.csv",
full.names = TRUE)
bin100_basenames <-
bin100_filelist %>%
str_remove("./data/") %>%
str_remove("_binsize100.csv")
#### 6.2 Raster plot ####
## For each unbinned file, generate a raster plot
rasterplots <- NULL
for (i in 1:length(unbinned_filelist)) {
## Read in the data
unbinned_data <-
read_csv(unbinned_filelist[i]) %>%
as_tibble()
## determine the max number of replicates
max_reps <- max(unbinned_data$Replicate)
## get unique speeds
sorted_speeds <-
unbinned_data$Speed %>% unique %>% sort(decreasing = TRUE)
## Generate the code for the ggplot and save it as rasterplots[[i]]
rasterplots[[i]] <-
unbinned_data %>%
## Remove any rows where spiking does not occur in the Spikes column
filter(Spikes == 1) %>%
## Convert Trial and Speed into factors and specify their level ordering
## This will make it easier to get the subplots in the order we want them
mutate(
Trial = factor(Trial,
levels = c("blank", "stationary", "moving")),
Speed = factor(Speed,
levels = sorted_speeds)) %>%
ggplot(aes(x = Time_stand, y = Replicate)) +
## The next three blocks will undershade each subplot according to stimulus
## phase (i.e., blank, stationary, moving)
annotate("rect",
xmin = 0, xmax = first(unbinned_data$Blank_end),
ymin = 0.5, ymax = max_reps + 0.5,
alpha = 0.075, color = NA, fill = "red") +
annotate("rect",
xmin = first(unbinned_data$Blank_end),
xmax = first(unbinned_data$Static_end),
ymin = 0.5, ymax = max_reps + 0.5,
alpha = 0.075, color = NA, fill = "darkgoldenrod1") +
annotate("rect",
xmin = first(unbinned_data$Static_end), xmax = 5,
ymin = 0.5, ymax = max_reps + 0.5,
alpha = 0.075, color = NA, fill = "forestgreen") +
## Up to 10 replicates were used, so we will force the y-axis to go to 10
scale_y_continuous(
limits = c(0.5, max_reps + 0.5),
expand = c(0, 0),
breaks = c(max_reps/2, max_reps)
) +
## There are multiple ways to plot a spike event. Since 100% of the rows in
## this filtered data set are spike events, we can simply plot a symbol at
## each time (Time_stand) that appears in the data. The `|` symbol is a good
## choice.
geom_point(pch = '|', size = 1.5) +
xlab("Time (sec)") +
ggtitle(paste0(unbinned_basenames[i], " raster")) +
## Use facet_grid() to create a grid of subplots. Rows will correspond to
## Speeds, and columns correspond to Directions
facet_grid(rows = vars(Speed), cols = vars(Direction)) +
theme_classic() +
theme(legend.position = 'none',
panel.spacing = unit(0.1, "lines"))
## Clean up
rm(unbinned_data)
}
#### 6.3 Mean spike rate plots ####
## Bin size = 100
## For each 100-ms binned file, generate a mean spike plot
bin100_msr_plots <- NULL
for (i in 1:length(bin100_filelist)) {
## Read in the data
bin100_data <-
read_csv(bin100_filelist[i]) %>%
as_tibble()
## get unique speeds
sorted_speeds <-
bin100_data$Speed %>% unique %>% sort(decreasing = TRUE)
## Compute SE and other metrics and add this to our data set
dataslices_100 <-
bin100_data %>%
mutate(
Speed = factor(Speed,
levels = sorted_speeds)) %>%
## Split by direction and speed, because we will use those to define each
## subplot
group_split(Direction, Speed) %>%
## Group by time bin
purrr::map(group_by, bin) %>%
## Within each time bin, compute the following:
purrr::map(transmute,
## first() can be used for metadata such as Speed or Direction
Speed = first(Speed),
Direction = first(Direction),
## I generally compute the mean within each bin for the following:
Time_stand = mean(Time_stand),
Blank_end = mean(Blank_end),
Static_end = mean(Static_end),
Mean_spike_rate = mean(Spike_rate),
## To get SE, divide s.d. by the square root of sample size
Spike_rate_SE = sd(Spike_rate)/sqrt(n()),
Mean_photod_rate = mean(Photod_mean),
## SE of photodiode
Photod_SE = sd(Photod_mean)/sqrt(n())
) %>%
purrr::map(ungroup) %>%
bind_rows()
## Generate the mean spike rate plot using ggplot
bin100_msr_plots[[i]] <-
dataslices_100 %>%
## The same code block can be used to generate either the mean spike rate
## (shown below) or photodiode trace (commented out)
ggplot(aes(x = Time_stand,
y = Mean_spike_rate #Mean_photod_rate
)) +
## We'll actually start by placing red, yellow, and green vertical lines to
## distinguish between blank, stationary, and moving phases
## This comes first so that it is the bottom-most layer and doesn't obstruct
## the data
geom_vline(xintercept = 0, col = "red") +
geom_vline(xintercept = first(dataslices_100$Blank_end),
col = "darkgoldenrod1") +
geom_vline(xintercept = first(dataslices_100$Static_end),
col = "forestgreen") +
## We'll use `geom_ribbon()` to shade in the SE traces
geom_ribbon(aes(
ymin = Mean_spike_rate - Spike_rate_SE,
ymax = Mean_spike_rate + Spike_rate_SE
# ymin = Mean_photod_rate - Photod_SE,
# ymax = Mean_photod_rate + Photod_SE
),
fill = "grey80") +
## `geom_line()` will be used to draw the mean spike rate itself on top of
## the SE traces
geom_line(linewidth = 0.05) +
## Add a title to help us know what cell this is
ggtitle(bin100_basenames[i], " mean spike rate (100-ms bins)") +
xlab("Time (sec)") +
ylab("Spike rate (spikes/sec)") +
## To sub-plot by Speed and Direction, I typically use `facet_grid()`. This
## method allows me to explicitly declare what the row- and column-wise
## grouping variables are
facet_grid(rows = vars(Speed), cols = vars(Direction)) +
theme_classic()
rm(bin100_data, dataslices_100)
}
#### 6.2.1 Export to PDF ####
## Use the `pdf()` function to start the graphics device driver for producing
## PDFs
## Aspects such as page size and centering mode can be adjusted
for (i in 1:length(rasterplots)) {
pdf(file =
paste0("./plot_pdfs/",
unbinned_basenames[i],
"_raster.pdf"),
width = 22, height = 12,
pagecentre = TRUE, colormodel = "srgb")
## Now add the plot to the PDF simply by calling plot()
plot(rasterplots[[i]])
## To declare an end to this PDF writing session, use `dev.off()`
dev.off()
}
for (i in 1:length(bin100_msr_plots)) {
pdf(file =
paste0("./plot_pdfs/",
bin100_basenames[i],
"_raster.pdf"),
width = 22, height = 12,
pagecentre = TRUE, colormodel = "srgb")
plot(bin100_msr_plots[[i]])
dev.off()
}
#### 7 Polar direction tuning plots ####
#### 7.1 Data import and baseline rate measurement ####
bin10_data <- NULL
polar_directions <- NULL
baseline_summary_df <- NULL
for (i in 1:length(bin10_filelist)) {
## Read in the data
bin10_data[[i]] <-
read_csv(bin10_filelist[i], show_col_types = FALSE) %>%
as_tibble()
## get unique speeds
sorted_speeds <-
bin10_data[[i]]$Speed %>% unique %>% sort(decreasing = TRUE)
## Again, we will set Speed as an ordered factor to help control plotting
## later on
bin10_data[[i]] <-
bin10_data[[i]] %>%