%% Tailed Primer Design
% % The goal of this script is to take an input file containing untailed RPA 
% Basic primers and generate tailed primer sets to screen

%% Import data and labels

% Import CSV containing untailed RPA Basic primer sequences
% CSV has 3 columns: Target, Forward Primer, Reverse Primer
filename = 'nontailedsequences.csv';

% Read data in file into a table
datatable = readtable(filename);

% Convert table to cell array
primerdesign = table2struct(datatable);


%% Store properties of the original primers in structure field

for i=1:length(primerdesign)
    primerdesign(i).forwardprimerprops = oligoprop(primerdesign(i).ForwardPrimer);
    primerdesign(i).reverseprimerprops = oligoprop(primerdesign(i).ReversePrimer);
end

%% Generate random 15bp sequence tails

% Generate j random sequences, and attach in new column in cell
numtails =200;
i=1; 
j=1;

for i = 1:length(primerdesign) %number of HPV types in imported table
    for j= 1:numtails %number of tail sequences to generate
        %create an array with the first 15 bp tail
        if j == 1 
            randtail = randseq(15);
            randtails = [randtail];
        %For subsequent tails after the first one, add new tail to the array
        else
            randtail = randseq(15); 
            randtails = [randtails; randtail];
        end
    end
    %store this random tail array in the primerdesign structure
    primerdesign(i).tails = randtails;
end


%% Generate tailed forward and reverse primers

% Reset counters
i=i;
j=i;

% Cycle through the number of rows
for i = 1:length(primerdesign)
    % Step through the number of tails
    for j= 1:numtails
        % Create tailed forward and reverse primers
        primerdesign(i).tailedforward(j,:) = strcat(primerdesign(i).tails(j,:), primerdesign(i).ForwardPrimer);
        primerdesign(i).tailedreverse(j,:) = strcat(primerdesign(i).tails(j,:), primerdesign(i).ReversePrimer);
        
        % Store a separate set of tailed primers that
        %have internal spacer between tail and primer for ease of ordering
        primerdesign(i).tailedforwardspaced(j,:) = strcat(primerdesign(i).tails(j,:), '/iSpC3/' ,primerdesign(i).ForwardPrimer);
        primerdesign(i).tailedreversespaced(j,:) = strcat(primerdesign(i).tails(j,:), '/iSpC3/', primerdesign(i).ReversePrimer);
        
        % Store the oligonucleotide properties of tailed forward/reverse primers 
        primerdesign(i).tailprops(j,1)= oligoprop(primerdesign(i).tails(j,:));
        primerdesign(i).tailedforwardprops(j,1) = oligoprop (primerdesign(i).tailedforward(j,:));
        primerdesign(i).tailedreverseprops(j,1) = oligoprop(primerdesign(i).tailedreverse(j,:));
    end
end


%% Filter Bad Tails and Bad Tailed Primers 
% Code based on Exploring Primer Design Help Page from MathWorks
% https://www.mathworks.com/help/bioinfo/examples/exploring-primer-design.html
i=1;
j=1;

for i = 1:length(primerdesign)
    
    % (i) Find tails with GC content greater than 45%
    primerdesign(i).badtailsgc = [primerdesign(i).tailprops.GC]' > 45 ;
    
    for j=1:numtails
        %(iiA): Find tailed primers with MORE dimers than the origianl
        % primers
         primerdesign(i).badfpdimers(j,1) = size(primerdesign(i).tailedforwardprops(j).Dimers,1) > size(primerdesign(i).forwardprimerprops.Dimers,1);
         primerdesign(i).badrpdimers(j,1) = size(primerdesign(i).tailedreverseprops(j).Dimers,1) > size(primerdesign(i).reverseprimerprops.Dimers,1);

         % (iiB): Find tailed primers with MORE hairpins than the original
         % primers
         primerdesign(i).badfphairpins(j,1) = size(primerdesign(i).tailedforwardprops(j).Hairpins,1) > size(primerdesign(i).forwardprimerprops.Hairpins,1);
         primerdesign(i).badrphairpins(j,1) = size(primerdesign(i).tailedreverseprops(j).Hairpins,1) > size(primerdesign(i).reverseprimerprops.Hairpins,1);
    end
end

%% Combine'bad' arrays to find good tailed forward or reverse primers
i = 1;
for i=1:length(primerdesign)
    %Storing all the bad primers into one array
    
    primerdesign(i).badtailedfpsall = [primerdesign(i).badtailsgc, primerdesign(i).badfpdimers, primerdesign(i).badfphairpins];
    primerdesign(i).badtailedrpsall = [primerdesign(i).badtailsgc, primerdesign(i).badrpdimers, primerdesign(i).badrphairpins];
    
    % At this point, '1' indicates a tailed primer that fails screening criteria
    
    % Switch '1's to be good primers and then scan through each row to see
    % if a tailed primer meets all the criteria and store the position of
    % tailed primers that pass screening
    primerdesign(i).goodfppos = find(all(~primerdesign(i).badtailedfpsall,2));
    primerdesign(i).goodrppos = find(all(~primerdesign(i).badtailedrpsall,2));
    
    %Pull out the tailed primers that pass 
    primerdesign(i).goodtailedfps = primerdesign(i).tailedforward(primerdesign(i).goodfppos, :);
    primerdesign(i).goodtailedrps = primerdesign(i).tailedreverse(primerdesign(i).goodrppos, :);
    
    % Pull out the corresponding tailed primers with spacers
    primerdesign(i).goodtailedfpsspaced = primerdesign(i).tailedforwardspaced(primerdesign(i).goodfppos, :);
    primerdesign(i).goodtailedrpsspaced = primerdesign(i).tailedreversespaced(primerdesign(i).goodrppos, :);
end


%% Align tailed primers and checking for additional crossdimerization

% Check for cross dimerization
scr_mat = [-1,-1,-1,1;-1,-1,1,-1;-1,1,-1,-1;1,-1,-1,-1;];

i=1;
j=1;
k=1;

for i=1:length(primerdesign)
    % Set the Cross-Dimerization Threshold as the Dimer threshold
    % between untailed forward/reverse primers
    primerdesign(i).dimerthreshold = swalign(primerdesign(i).ForwardPrimer, primerdesign(i).ReversePrimer, ...
      'SCORINGMATRIX', scr_mat, 'GAPOPEN',5,'ALPHA','NT');
  
  % Loop through the number of good tailed forward primers and storing their
  % dimerization with the reverse primer
  if size(primerdesign(i).goodtailedfps, 1) >0  
    for j = 1: size(primerdesign(i).goodtailedfps, 1)
      primerdesign(i).crossdimer_tfp(j) = swalign(primerdesign(i).ReversePrimer,...
          primerdesign(i).goodtailedfps(j,:), 'SCORINGMATRIX', scr_mat,'GAPOPEN',5,...
          'ALPHA','NT');
    end
  end
   
  % Loop through the number of good tailed reverse primers and storing their
  % dimerization score with the forward primer
  if size(primerdesign(i).goodtailedrps, 1) >0
    for k=1:size(primerdesign(i).goodtailedrps, 1)
        primerdesign(i).crossdimer_trp(k) = swalign(primerdesign(i).ForwardPrimer,...
            primerdesign(i).goodtailedrps(k,:), 'SCORINGMATRIX', scr_mat, 'GAPOPEN',5,...
            'ALPHA','NT');
    end
  end

  % (iii) Find tailed primer combos that do not result in additional
  % cross-dimerization compared to the untailed primer pair
primerdesign(i).finaltfps_pos = find(primerdesign(i).crossdimer_tfp <= primerdesign(i).dimerthreshold);
primerdesign(i).finaltrps_pos = find(primerdesign(i).crossdimer_trp <= primerdesign(i).dimerthreshold);

% These are the final sequences to input into IDT containing the C3 spacer
primerdesign(i).finaltfpsspaced = primerdesign(i).goodtailedfpsspaced(primerdesign(i).finaltfps_pos,:);
primerdesign(i).finaltrpsspaced = primerdesign(i).goodtailedrpsspaced(primerdesign(i).finaltrps_pos, :);

end