-
Notifications
You must be signed in to change notification settings - Fork 1
/
kilosort_preprocess_forcluster.m
213 lines (185 loc) · 6.34 KB
/
kilosort_preprocess_forcluster.m
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
%%
function kilosort_preprocess_forcluster(varargin)
% ---------------------
% written by Emily Jane Dennis 2020-06-14 (during pandemic!)
% Edited by Jess Breda 2020-07 (just literally all july)
% purpose is to take in binary files for each tetrode bundle from a 128
% channel (32 tetrode) drive and zero giant sections where
% there is huge noise that we want to ignore. We also want to output some
% plots and save them so we can do some quick validation by eye.
%
%
% TODO:
% - clean up ifelse checks for inputs
% - add more plots to outputs
% - come back after some kilosort testing and add a if/else statement for
% if we should be saving the new binary files with the bad sections
% removed, zerod, or interpd
% - make threshold and window optional inputs
%
% INPUT PARAMETERS:
% - none needed - it should be able to be run from a local folder
%
% OPTIONAL PARAMETERS:
% - string directing us to a folder
% - chan = number of channels in the binary file, defaults to 32
% - ops.fs - param for butterworth filter
% - ops.fshigh - param for butterworth filter
%
% RETURNS:
% - nothing, but saves a new binary file and outputs several plots for
% quality control
%
% = EXAMPLE CALLS:
% -
% ---------------------
%% inputs
if isempty(varargin)
homedirectory=pwd;
threshold = 0.5;
window = 10000;
chan=32;
ops.fs = 32000;
ops.fshigh = 300;
%if user provides input, check that it is a string for a directory
elseif length(varargin)==1
if isfolder(varargin{1})
homedirectory=varargin{1}; %#ok<NASGU>
cd(homedirectory)
else
error('input must be in the format of a string leading to a directory')
end
threshold = 0.5;
window = 10000;
chan=32;
ops.fs = 32000;
ops.fshigh = 300;
elseif length(varargin)==2
if isfolder(varargin{1})
homedirectory=varargin{1}; %#ok<NASGU>
cd(homedirectory)
else
error('input must be in the format of a string leading to a directory')
end
threshold = varargin{2};
window = 10000;
chan=32;
ops.fs = 32000;
ops.fshigh = 300;
elseif length(varargin) == 3
if isfolder(varargin{1})
homedirectory=varargin{1}; %#ok<NASGU>
cd(homedirectory)
else
error('input must be in the format of a string leading to a directory')
end
threshold = varargin{2};
window = varargin{3};
chan=32;
ops.fs = 32000;
ops.fshigh = 300;
elseif length(varargin) == 4
if isfolder(varargin{1})
homedirectory=varargin{1}; %#ok<NASGU>
cd(homedirectory)
else
error('input must be in the format of a string leading to a directory')
end
threshold = varargin{2};
window = varargin{3};
chan = varargin{4};
ops.fs = 32000;
ops.fshigh = 300;
elseif length(varargin) == 5
if isfolder(varargin{1})
homedirectory=varargin{1}; %#ok<NASGU>
cd(homedirectory)
else
error('input must be in the format of a string leading to a directory')
end
threshold = varargin{2};
window = varargin{3};
chan = varargin{4};
ops.fs = varargin{5};
ops.fshigh = 300;
else
if isfolder(varargin{1})
homedirectory=varargin{1}; %#ok<NASGU>
cd(homedirectory)
else
error('input must be in the format of a string leading to a directory')
end
threshold = varargin{2};
window = varargin{3};
chan = varargin{4};
ops.fs = varargin{5};
ops.fshigh = varargin{6};
end
%% filter & initialization
% make a filter for the data based on ops inputs/defaults
% butterworth filter with only 3 nodes (otherwise it's unstable for float32)
% Wn (ops.fshigh/ops.fs) is the cutoff frequency, and must be between 0.0 and 1.0. 1.0 is half the
% sample rate
% high means it's a highpass filter
%outputs filter coefficients b1 (numerator) and a1 (denominator)
[b1, a1] = butter(3, ops.fshigh/(ops.fs/2), 'high');
% make list of files to process
listofbinaryfiles=dir('*.bin');
%make empty fftosave
fftosave=[];
%determine computer type
if ispc
delim='\';
else
delim='/';
end
%% loop over binary files
for i = 1:length(listofbinaryfiles)
% first, open the binary file to read
fname = listofbinaryfiles(i).name;
fid=fopen(fname,'r');
% next, name and open a new binary file to write to, and put it in it's
% own folder
mkdir(fullfile(homedirectory, delim, sprintf('%s_T%s_W%s_forkilosort%s',fname(1:end-4), num2str(threshold*10), num2str(window))))
kilosortbinfolder = [homedirectory, delim, sprintf('%s_T%s_W%s_forkilosort',fname(1:end-4), num2str(threshold*10), num2str(window))]
addpath(kilosortbinfolder)
cd(kilosortbinfolder)
fidw = fopen(sprintf('%s_T%s_W%s_forkilosort.bin',fname(1:end-4), num2str(threshold*10), num2str(window)), 'w');
while 1
% now, read in a PORTION of the data. Format it as a matrix with chan rows and
% 1e5 values - we will loop through this for each file until all data is
% read in
dataRAW = fread(fid, [chan 1e6], 'int16');
sizeofdata=size(dataRAW);
if sizeofdata(2) == 0
break %breaks the while loop
end
% transpose
dataRAW = dataRAW';
% divide by 1000 because the filter prefers that
dataRAW = double(dataRAW)/1000;
% apply the filter
datr = filtfilt(b1, a1, dataRAW);
dataFILT = datr; %renaming now so I can plot later without overwriting
% make a binary 'mask' for the data, looking for big changes to
% remove
% first we want absolute values, so we square datr, take the means of each
% row (channel), and then take the square root
dataABSVAL = mean(datr.^2, 2).^.5;
% create a binary mask where dataABSVAL > threshold
% then, take moving mean of mask with window size
mask1MEAN = movmean(double(dataABSVAL>threshold), window);
% binary mask 'signal' = 1, 'noise' = 0
mask2= mask1MEAN < 0.00001;
% mask data & set noise to 0
dataMASK = datr .* mask2;
% save
dat16 = int16(1000*dataMASK');
fwrite(fidw, dat16, 'int16');
end
fclose(fid);
fclose(fidw);
sprintf('finished file %d of %d files to process',i,length(listofbinaryfiles))
cd .. %return to directory with other binary files
end
end