How to count the number of elements of a cell having string and integers?

Hello guys,

I have a cell array of strings and integer elements as shown here:

'POLR3D,SNAPC5'

'HIST3H2BB,HIST3H2BB'

'TFAM'

0

'CYP4F2'

0

'TRNT1,NUP214,NUP160,TRNT1'

'RPS21'

'ACTR3'

0

I used this code to count the elements of this cell but doesn't work , i think because this cell have multiple types.

So is it possible to count this cell elements or need some transformations types as i tried int2str, string but non of them worked!

SS=(tt(:,2));
n3=size(SS,1);
  sub_stage_col2 = cell(1, n3);
   for k1 = 1:n3
 sub_stage_col2{k1} = strtrim(strsplit(SS{k1,:} ,','));
   end
  ptws_count2 = cellfun(@(x) numel(x), sub_stage_col2);
 for m1=1:n3
    pathws_count2(m1,1)=ptws_count2(m1);
 end
 

Also this cell have some repeated elements in some of its rows in rows 2 and 7: 'HIST3H2BB,HIST3H2BB', 'TRNT1,NUP214,NUP160,TRNT1 'How to remove the duplicates ?

I again appreciate any help!

[Image Analyst]

Help us to help you.

Attach your cell array (is it tt?) in a .mat file with the paper clip icon.

[chocho]

@Image Analyst ok wait one minute i will share it

I have soved the issue of duplications now i just want to count the elements Thanks

[chocho]

@Image Analyst I want to count the elements of this cell array and when there is 0 put it 0 bcz i mean by 0 no element.

I attached the matfile

[chocho]

Can anyone help me out plz ?

Answer 1

[TADA]

because you have multiple types, most likely you will need to either selectively change to a uniforrm data type at the begining or handle the data selectively according to it's datatype (i.e using ischar/isnumeric)

I assumed the cell array you supplied is this:

{'POLR3D,SNAPC5'; 'HIST3H2BB,HIST3H2BB'; 'TFAM'; 0; 'CYP4F2'; 0; 'TRNT1,NUP214,NUP160,TRNT1'; 'RPS21'; 'ACTR3'; 0};

and I assumed it's the value of your SS variable. historically speaking, this is an ill name choice if you ask me. That being said, I got this going:

SS = {'POLR3D,SNAPC5'; 'HIST3H2BB,HIST3H2BB'; 'TFAM'; 0; 'CYP4F2'; 0; 'TRNT1,NUP214,NUP160,TRNT1'; 'RPS21'; 'ACTR3'; 0};
n3 = size(SS,1);
sub_stage_col2 = cell(1, n3);

for k1 = 1:n3
    if ischar(SS{k1})
        sub_stage_col2{k1} = unique(strtrim(strsplit(SS{k1} ,',')));
    else
        sub_stage_col2{k1} = SS{k1};
    end
end

pathws_count2(:,1) = cellfun(@(x) numel(x), sub_stage_col2(:));

disp(pathws_count2);

     2
     1
     1
     1
     1
     1
     3
     1
     1
     1

the use of unique function gets rid of the duplicates within each row

Putting the result of cellfun in one array then iterating to copy to a second array is unnecessary, even if the number of rows in pathws_count2 and sub_stage_col2 is different (then you can set only the desired rows.

and the call to strsplit is a potential bug:

strsplit(SS{k1,:}, ',')

will be the same as calling

strsplit(SS{k1}, ',')

unless SS is a matrix with more than one column (or a row vector), in which case, you will probably get an exception since it's like calling

strsplit(SS{k1,1}, SS{k1,2}, SS{k1,3},...,',')

[chocho]

@TADA Tahnks for your answer but when there is 0 i want to be 0 because 0 means no element is there is it possible ?

Also i want to output another cell wiich remove duplicates and keep the others the questions have 2 separate parts one part count the number of elements of every cell and the other remove duplicates and output in another cell i see you do both altogether

Thanks

[TADA]

You Can Save The Output Cell Array Without unique Then Invoke unique And Save That In Another Cell Array.

As For The Zero, Just Change

% Not  Tested 

if isnumeric(SS{k1}) 
    if SS{k1} == 0 
sub_stage_col2{k1} = '';
    else 
        % If You Need It 
        sub_stage_col2{k1} = SS{k1}; 
    end
end

I Don't Know Where You Get That Data From But Perhaps You Can Change The Way You Parse That Data To begin With To A Better Data structure

[chocho]

@TADA Thanks TADA for your great help!