Extracting sequential and nonsequential list of scores from cell array

Hello all,

I have a list of scores where I want to extract the Wake scores and then also the NREM scores (any combination of N1, N2, N3). I only want to extract them if there are more than 7 epochs in the list of scores. I would also like the take the values if there are 7 epochs that are sequential (e.g. test 3).

When I run the following code I get the correct answer:

valid_sequences =     2     9

scores = {'N1' ' N3' 'Wake' 'Wake' 'Wake' 'Wake' 'Wake' 'Wake' 'Wake' 'N1'};

sleepStages = {'Wake', 'NREM'};
sleepStagesMap = containers.Map({'Wake', 'N1', 'N2', 'N3'}, [1 2 2 2]);
sequence_start = 1;
num_scores = numel(scores);
valid_sequences = [];

for j = 1:num_scores
    stage = scores{j};
    
    if strcmp(stage, 'Wake')
        if j == num_scores
            sequence_length = j - sequence_start + 1;
        else
            sequence_length = j - sequence_start;
        end
        
        if sequence_length >= 7
            valid_sequences = [valid_sequences; sequence_start, j];
        end
    elseif ismember(stage, {'N1', 'N2', 'N3'})
        sequence_length = j - sequence_start;
        
        if sequence_length >= 7 && sum(ismember(scores(sequence_start:j), {'N1', 'N2', 'N3'})) >= 7
            valid_sequences = [valid_sequences; sequence_start, j];
        end
    else
        sequence_start = j;
    end
end

However, when I run the same code with the following tests I get incorrect answers:

% test 2
scores = {'Wake' 'Wake' 'N1' 'N1' 'N2' 'N3' 'N2' 'N2' 'N3' 'Wake'};

valid_sequences =
1     9
1    10

% test 3
scores = {'Wake' 'Wake' 'Wake' 'Wake' 'Wake' 'Wake' 'N2' 'Wake' 'Wake' 'Wake'};

valid_sequences =
1     9
1    10

Can you help me? I've been stuck on this for a couple of days, and just keep running into the same answers. Thanks so much.

[Menika]

Hi,

Can you confirm what results are you expecting for test 2 and 3 in valid_sequence?

[nines]

they should be:

test 2: 2 9

test 3: 1 6; 7 10

I am going to use the epochs to extract corresponding timeseries.

Ideally, it would be nice to have the following because I think it is less confusing:

test 1: 3 9

test 2: 3 9

test 3: 1 6; 8 10