synced with private copy

IO/load_smf.m

@@ -0,0 +1,36 @@
+function [file_smfit, file_smfit_std] = load_smf(smfitnessfile, sgadata, lfid)
+% function [file_smfit, file_smfit_std] = load_smf(smfitnessfile, sgadata, lfid)
+
+   smf_fid = fopen(smfitnessfile, 'r');
+   fitness_data = struct();
+   fitness_data.raw = textscan(smf_fid, '%s%f%f', 'Delimiter', '\t', 'ReturnOnError', false);
+   fclose(smf_fid);
+   fitness_data.ORF = fitness_data.raw{1};
+   fitness_data.SMF = fitness_data.raw{2};
+   fitness_data.STD = fitness_data.raw{3};
+   fitness_report_header = {'Exact match', 'Partial Match', 'Not Found', 'NaN in file'};
+   fitness_report_counts = zeros(1,4);
+   fitness_hash = hash_strings(fitness_data.ORF);
+
+   file_smfit = zeros(length(sgadata.orfnames),1) + NaN;
+   file_smfit_std = zeros(length(sgadata.orfnames),1) + NaN;
+   for i=1:length(sgadata.orfnames)
+       if fitness_hash.containsKey(sgadata.orfnames{i})
+           file_smfit(i)     = fitness_data.SMF(fitness_hash.get(sgadata.orfnames{i}));
+           file_smfit_std(i) = fitness_data.STD(fitness_hash.get(sgadata.orfnames{i}));
+           fitness_report_counts(1) = fitness_report_counts(1)+1;
+       elseif fitness_hash.containsKey(strip_annotation(sgadata.orfnames{i}, 'last'))
+           file_smfit(i)     = fitness_data.SMF(fitness_hash.get(strip_annotation(sgadata.orfnames{i}, 'last')));
+           file_smfit_std(i) = fitness_data.STD(fitness_hash.get(strip_annotation(sgadata.orfnames{i}, 'last')));
+           fitness_report_counts(2) = fitness_report_counts(2)+1;
+       else
+           fitness_report_counts(3) = fitness_report_counts(3)+1;
+       end
+   end
+   fitness_report_counts(4) = sum(isnan(file_smfit)) - fitness_report_counts(3);
+
+   log_printf(lfid, 'Fitness file report:\n');
+   for i=1:length(fitness_report_header)
+       log_printf(lfid, '\t%s\t: %d\n', fitness_report_header{i}, fitness_report_counts(i));
+   end
+end

IO/output_interaction_data.m

@@ -1,18 +1,18 @@
 %%
-% OUTPUT_INTERACTION_DATA - prints out the calculated genetic interactions into a tab delimited text file.
+% OUTPUT_INTERACTION_DATA - prints out the calculated genetic interactions 
+% into a tab delimited text file.
 %
 %
-% Inputs:
+% Authors: Chad Myers ([email protected]), 
+% Anastasia Baryshnikova ([email protected])
+% Benjamin VanderSluis ([email protected])
 %
-% Authors: Chad Myers ([email protected]), Anastasia Baryshnikova ([email protected])
-%
-% Last revision: 2010-07-19
+% Last revision: 2017-12-12
 %
 %%
 
-function output_interaction_data(outputfile,orfnames,escores,escores_std,background_mean,background_std,smfit,smfit_std,dm_expected,dm_actual,dm_actual_std,lfid)
-
-pvals = sqrt(normcdf(-abs(escores./escores_std)) .* normcdf(-abs(log((background_mean + escores)./background_mean) ./ log(background_std) )));
+function output_interaction_data(outputfile, orfnames, escores, escores_std, ...
+   pvals, smfit, smfit_std, dm_expected, dm_actual, dm_actual_std, lfid)
 
 log_printf(lfid, ['Printing output file...\n|' blanks(50) '|\n|']);
 
@@ -21,26 +21,26 @@ function output_interaction_data(outputfile,orfnames,escores,escores_std,backgro
     for j = 1:length(escores)
         if ~isnan(escores(i,j))
               fprintf(fid,'%s\t%s\t%f\t%f\t%e\t%f\t%f\t%f\t%f\t%f\t%f\t%f\n', ...
-                  orfnames{i}, orfnames{j}, escores(i,j), escores_std(i,j), pvals(i,j), ...
+                  orfnames{i}, orfnames{j}, escores(i,j), escores_std(i,j), ...
+                  pvals(i,j), ...
                   smfit(i), smfit_std(i), smfit(j), smfit_std(j), ...
                   dm_expected(i,j), dm_actual(i,j), dm_actual_std(i,j));
         end
     end
 
-    % Print progress
     print_progress(lfid, length(escores), i);
 
 end
 fclose(fid);
 
-% only output an orf if it has scores
-% bad arrays are appearing otherwise
+% output the list of strains, but only if they have scores
+% otherwise bad arrays appear
 valid_queries = orfnames(sum(~isnan(escores),2)>0);
 valid_arrays  = orfnames(sum(~isnan(escores),1)>0);
 valid_orfs = union(valid_queries, valid_arrays);
 fid = fopen([outputfile '.orf'], 'w');
 for i=1:length(valid_orfs)
-	fprintf(fid, '%s\n', valid_orfs{i});
+   fprintf(fid, '%s\n', valid_orfs{i});
 end
 fclose(fid);
 

README.md

@@ -78,7 +78,7 @@ SAFE-ENTRY points
 The script has a number of places marked:
 `% SAFE ENTRY`
 Which means "The only variables which change after this point, are created after this point."
-These have been determined by manual inspection and are by no means exhaustive. They are used to resme
+These have been determined by manual inspection and are by no means exhaustive. They are used to resume
 the script mid-execution to save time when testing changes. I generally use this procedure:
   - Locate the position of the change you wish to test (line XXX)
   - Locate the first SAFE ENTRY above XXX, (line YYY)