neuro_cbs/example_train.m at master · pesaranlab/neuro_cbs · GitHub

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
%%%%%
%%%%% This script shows process of applying Classification-Based
%%%%% Selection (CBS) to a set of densely sampled bank recordings from
%%%%% a Neuropixel Phase 1 probe.
%%%%%

clear;

%%%%%
%%%%% First assemble a cell array of file paths to each dense bank
%%%%% recording. Recordings are assumed to be high-pass filtered and sorted
%%%%% by Kilosort2.
%%%%%

raw_files = {...
    '../data/rec_bank0/rec_bank0_imec0/rec_bank0_t0.imec0.ap.bin', ...
    '../data/rec_bank1/rec_bank1_imec0/rec_bank1_t0.imec0.ap.bin', ...
    '../data/rec_bank2/rec_bank2_imec0/rec_bank2_t0.imec0.ap.bin', ...
    };

%%%%%
%%%%% Then, specify options. See inside functions below for more options.
%%%%%

ops.selection_method = 'cbs';  % possible choices: {'cbs', 'mucbs'}
ops.nspikes_per_unit = 100;    % to match paper, set to 100


%%%%%
%%%%% Run preprocessing on individual bank recordings
%%%%%

bank_models = get_bank_model_neuropixel(raw_files, ops);

%%%%%
%%%%% optimize selection map
%%%%%

ops.do_classifier = true; % runs a classification validation every pass.
                           % At the end of optimization, the selection
                           % map corresponding with the highest validation
                           % accuracy is chosen. Sometimes this is not the
                           % terminal selection map. If set to false,
                           % CBS skips the validation and just returns the
                           % terminal selection map. In the paper, we used
                           % true.
ops.banks_to_validate = 1:2;   % restricts classification validation to
                               % deepest 2 banks, as used in the paper.
ops.do_plot = true;            % plot optimization summary
ops.save_plot = true;          % save plot summary
ops.plot_dir = '../results'; % plot save directory

ops.init_method = 'mucbs';
selection_map = optimize_selection_map(bank_models, ops);

%%%%%
%%%%% write output map to an .imro file
%%%%%

imro_file_path = '../results/cbs_map.imro';
write_cbs_imro_file(imro_file_path, selection_map, bank_models);

fprintf('\n\nSee results directory for optimization summary plots\n');