Merging the unique values

final_result = 

      'Genes'      'T0&T2'    'T1&T3'    'T2&T4'    'T3&T5'    'T4&T6'
      'YAR029W'    'dd'       'uu'       'dd'       'uu'       'du'   
      'YBL095W'    'du'       'ud'       'ud'       'du'       'du'   
      'YBL111C'    'uu'       'uu'       'ud'       'ud'       'du'   
      'YBL113C'    'uu'       'uu'       'uu'       'ud'       'ud'   
      'YBR096W'    'uu'       'uu'       'ud'       'ud'       'dd'   
      'YBR138C'    'ud'       'ud'       'ud'       'du'       'du'   

In these i have to find the genes having same variable(1st variable in each interval) for all the intervals for example

'YAR029W'has 'd'(1st variable) in interval 'T0&T2' ,in same interval i want to find which genes havibg variable'd',for example 

'YBL095W','YBR138C',has value 'd'('d' may be 1sy or 2nd variable ) 

Same should be done for 2nd gene also in interval 'T0&T2'      

'YBL111C' has 1st value 'u',the genes having 'u' must be displayed where ,    'YBL113C''YBR096W','YBR138C' must be displayed 

So i need output as 

'Gene1'      'T0&T2'

'YAR029W'    'dd'
'YBL095W'    'du'
'YBR138C'    'ud'
;
;
 'Gene3'      'T0&T2'
'YBL111C'    'uu'
'YBL113C'    'uu'
'YBR096W'    'uu'
'YBR138C'    'ud' 
;

;
I have 1000 genes please help

Answer 1

[Azzi Abdelmalek]

%i'm not sur if i've understood what you meant, but try this

 a=final_result;[n,m]=size(a);B=a(2:end,1);
for k=2:m
    A=a(2:end,k); C=[B A];
    ent1={strcat('out',num2str(2*k-3))  char(a(1,k))};
    ent2={strcat('out',num2str(2*k-2))  char(a(1,k))};
    s{1,k-1} =[ ent1;C(cellfun(@(x)any(regexp(x,'d')),A),:)];
    s{2,k-1}= [ ent2;C(cellfun(@(x)any(regexp(x,'u[u,d]')),A),:)];
    disp(s{1,k-1});disp(s{2,k-1})
end

Answer 2

[Andrei Bobrov]

EDIT

   final_result = {...
  'Genes'      'T0&T2'    'T1&T3'    'T2&T4'    'T3&T5'    'T4&T6'
  'YAR029W'    'dd'       'uu'       'dd'       'uu'       'du'
  'YBL095W'    'du'       'ud'       'ud'       'du'       'du'
  'YBL111C'    'uu'       'uu'       'ud'       'ud'       'du'
  'YBL113C'    'uu'       'uu'       'uu'       'ud'       'ud'
  'YBR096W'    'uu'       'uu'       'ud'       'ud'       'dd'
  'YBR138C'    'ud'       'ud'       'ud'       'du'       'du'  };

fr = final_result(2:end,1:2);
frd = fr(cellfun(@(x)any(ismember(x,'d'),2),fr(:,2)),:);
fru = fr(cellfun(@(x)strcmp(x(1),'u'),fr(:,2)),:);
out = [arrayfun(@(x)frd([1,x:end],:),(2:size(fru,1)-1)','un',0);...
       arrayfun(@(x)fru([x:end],:),(1:size(fru,1)-1)','un',0)];

[Pat]

Andrei as per your code i get output as 

out = 

      'YAR029W'    'dd'
      'YBL095W'    'du'
      'YBR138C'    'ud'
      'YBL111C'    'uu'
      'YBL113C'    'uu'
      'YBR096W'    'uu'
      'YBR138C'    'ud'

but i need output as 

out1(1st gene)
'YAR029W'    'dd'
'YBL095W'    'du' 
'YBR138C'    'ud'

out2

'YAR029W'    'dd'
'YBR138C'    'ud' 

out3
;
;

out4;
;
;

out5
'YBR096W'    'uu' 
'YBR138C'    'ud'

please help

[Jan]

I do not understand the format of your output. Should "out1", "out2" be separate variables? What are the "; ;" exactly - an [2 x 0] cell string?

[Andrei Bobrov]

Answer was edited.

Answer 3

[Azzi Abdelmalek]

combining with Andrei answer:

a=final_result;[n,m]=size(a);B=a(2:end,1);
for k=2:m
  A=a(2:end,k);
  s{1,k-1} = [ [char(a(1,k)) '/gen1']; B(cellfun(@(x)any(regexp(x,'d')),A),:)];
   s{2,k-1}= [[char(a(1,k))  '/gen3'] ;B(cellfun(@(x)any(regexp(x,'u[u,d]')),A),:)];
   disp(s{1,k-1});disp(s{2,k-1})
end

[Pat]

Azzi i cannot understand  the output of your code ,but i need the output as 

out1(1st gene)  'T0&T2'
'YAR029W'       'dd'
'YBL095W'       'du' 
'YBR138C'       'ud'

out2(2nd gene)

'YAR029W'       'dd'
'YBR138C'       'ud' 

out3
 'YBL111C'      'uu'         
 'YBL113C'      'uu'        
 'YBR096W'      'uu'    
'YBR138C'       'ud'

out4;
'YBL113C'       'uu'        
 'YBR096W'      'uu'          
 'YBR138C'      'ud'  

out5
'YBR096W'       'uu' 
'YBR138C'       'ud'

The 1st column is the Genes and the 2nd is Interval

Answer 4

[per isakson]

This code

    for ii = 2 : 6
        out = cell(0,2);
        first_character = final_result{ ii, 2 }(1);
        for jj = ii : 7
            if ismember( first_character, final_result{ jj, 2 } ) 
                out = cat( 1, out, final_result( jj, 1:2 ) ); 
            end    
        end
        out
    end

prints

out = 
    'YAR029W'    'dd'
    'YBL095W'    'du'
    'YBR138C'    'ud'
out = 
    'YBL095W'    'du'
    'YBR138C'    'ud'
out = 
    'YBL111C'    'uu'
    'YBL113C'    'uu'
    'YBR096W'    'uu'
    'YBR138C'    'ud'
out = 
    'YBL113C'    'uu'
    'YBR096W'    'uu'
    'YBR138C'    'ud'
out = 
    'YBR096W'    'uu'
    'YBR138C'    'ud'
>>

The columns

'T1&T3'    'T2&T4'    'T3&T5'    'T4&T6'

are confusing.

.

You write:

    out2(2nd gene)
    'YAR029W'       'dd'
    'YBR138C'       'ud'

is that in accordance with your requirements?

[Pat]

Thanks a lot per isakson, The columns

'T1&T3' 'T2&T4' 'T3&T5' 'T4&T6' has to be prossed another operation i have given out1,out2 because these has to be processed further for another operations

[per isakson]

I guessed that regarding the columns, but still not including them here would have made it easier to read the question.

You know, I cannot assign the result to out1, out2, etc. That would have made Jan and Walter very disappointed with me:) See the FAQ.