source: branches/uk_ac_ebi_Tab2MageImporter/t2m-wizard.pl @ 497

#!/usr/bin/perl 
# Include libraries
#author: rocca@ebi.ac.uk EMBL-EBI

#test command line:

#C:\Perl\workspace>perl t2m-wizard.pl  --org Homo_sapiens --array NuGO_hs --single 1 --trtgroup 6 --factors compound(aspirin,ibuprofen);dose(none,medium,high) --target liver,heart,brain --ref no --dye no --subject 4

#C:\Perl\workspace>perl t2m-wizard.pl  --org Homo_sapiens --array Agilent1 --single 2 --ref yes --dye no --trtgroup 6 --factors compound(aspirin,ibuprofen);dose(none,medium,high) --target liver,heart,brain --subject 4



use strict; # Disable automatic variables
use Fcntl;  # IMPORTANT: necessary for running sysopen function (flag , eg. O_WRONLY import)
use Getopt::Long;
use File::Find;
use File::Basename;
use File::Copy;

@ARGV=qw(.) unless @ARGV;

my $usage;

my $organism;
my $array;
my $trtgroup_nb;
my $subject_nb;
my $target_tissue;
my @target_tissue;
my $dyeswap;
#(yes/no);
my $reference;
# (yes/no);
my $single_or_multiple_ch;
# (single/double);
my $hyb_nb; 
#my $pooling; TO DO
#(yes/no);
#my @sampling_points;
my $record;
my @records;

my $factors;


my $tot_hyb_nb;

  GetOptions("organism=s"=>\$organism,
       "array=s"=>\$array,
       "single_or_multiple_channel=i"=>\$single_or_multiple_ch,   
       "trtgroup_nb=i"=>\$trtgroup_nb,
       "factors=s"=>\$factors,
       "subject_nb=i"=>\$subject_nb,
       "target_tissue=s"=>\$target_tissue,
       "reference=s"=>\$reference,
       "dyeswap=s"=>\$dyeswap);

  if ($usage    ||
      !$organism    ||
      !$array   ||
      !$single_or_multiple_ch ||
      !$factors   ||      
      !$trtgroup_nb   ||
      !$subject_nb  ||
      !$target_tissue ||  
      !$reference   ||
      !$dyeswap)
      
      { &usage; }


#need to implement input validation:


open(FILE, "+>t2m-wiz-output.txt") or die "cannot open \"t2m-wiz-output.txt\": $!";

#we create a default template for Experiment and Protocol Sections
#for user to modify manually and update accordingly


print FILE "Experiment section\n";
print FILE "domain\n";
print FILE "accession\n";
print FILE "quality_control\n";
print FILE "experiment_design_type\n";
print FILE "name\n";
print FILE "description\n";
print FILE "release_date\n";
print FILE "submission_date\n";
print FILE "submitter\n";
print FILE "organization\n";
print FILE "publication_title\n";
print FILE "authors\n"; 
print FILE "journal\n"; 
print FILE "volume\n";  
print FILE "issue\n"; 
print FILE "pages\n";
print FILE "year\n";
print FILE "pubmed_id\n\n";

print FILE "Protocol section\n";
print FILE "accession name  text  parameters  type\n";
print FILE "P-DIET-1  treatment diet    treatment\n";
print FILE "P-EXTR-1  extraction  RNA extraction    extraction\n";
print FILE "P-LABL-1  labeling  text    labeling\n";
print FILE "P-HYBR-1  hybridization text    hybridization\n";
print FILE "P-SCAN-1  scanning  text    scanning\n\n";

print FILE "Hybridization section\n";

# create the header of the hybridization section, a default core of mandatory fields.

#IMPORTANT NOTE: needs to modified if pooling is used


my $header= join (' ', "File[raw]",
"array[accession]",
"BioSource",
"BioMaterialCharacteristics[Organism]",
"BioMaterialCharacteristics[OrganismPart]",
"BioMaterialCharacteristics[CellType]",
"Sample",
"Protocol[grow]",
"Extract",
"Protocol[extraction]",
"LabeledExtract",
"Protocol[labeling]",
"Dye",
"Hybridization",
"Protocol[hybridization]",
"Scan",
"Protocol[scanning]",
"FactorValue[Treatment Type]");

print FILE "$header\n";






my @factors;
@factors = split(/;/, $factors);

@target_tissue = split(/,/ , $target_tissue);


print "species: $organism\n";
print "arraydesign: $array\n";
print "treatment groups: $trtgroup_nb\n";
print "subjects per group: $subject_nb\n";
print "tissue per subject: @target_tissue\n";




my $m;
for $m (0..$#factors) {

#need to use a hash -> more complex data structure

  if ($factors[$m]=~/\w+\(.*\)/) {
    my ($key, $values)=($factors[$m]=~/(\w+)\((.*)\)/);
    my @factorvalues=split(/,/, $values);
    my $number_of_values=$#factorvalues+1;
    print "factor:$key, $number_of_values associated values, which are: @factorvalues\n";
  }

}


#To IMPLEMENT:

#once the number of study groups is known, it would be good to get a full description for each study of the associated factors and factor levels:
#for each study group, indicate Factor name and intensity level, for example
#study group#1: [factor=Dose/Value=10 mg],[factor=Compound/Value=aspirin], [Factor=Duration/Value=24 hr]
#for each study group, identify the factor and their values
#for every organ derived from each animal in each study group


if ($single_or_multiple_ch == 1) { 

#considered here as equivalent to using Affymetrix platform


my $count=1;

for my $i (1..$trtgroup_nb) { #iterating through each studygroup/treatment group

  my $subjectcount=0; 
  #initializes a counter in order to provide unique identifier to study subject
  
  
  for  my $j (1..$subject_nb) { #iterating through each animal/patient/plant/culture of a treatment group (biological replicates for the same conditions)
  
    $subjectcount=($i-1)*$subject_nb+$j;
    #computes the actual number of subjects by summing up the current rank of a subject in a given study group to the number of subjects already created
  
    for  my $k (0..$#target_tissue) { #iterating through each tissue derived from each subject
    
    #we create a canonical tab2mage hybridization record

    my $tissue=$target_tissue[$k];
    $k++;

    $record=join('  ',
    "file-".$count.".CEL",
    $array, 
    "group-".$i."-subject-".$subjectcount,
    $organism,
    $tissue,
    $cellmodel,
    "group-".$i."-subject-".$subjectcount.".sample-".$k,
    "P-DIET-1",
    "group-".$i."-subject-".$subjectcount.".sample-".$k."-extract",
    "P-EXTR-1",
    "group-".$i."-subject-".$subjectcount.".sample-".$k."-extract-"."le",
    "P-LABL-1",
    "biotin",
    "hybridization-".$count,
    "P-HYBR-1",
    "scan-".$count,
    "P-SCAN-1",
    "trt_group".$i);
    
    #print "record: $record\n";

    push(@records,$record);
    
    $count++;
    
    }
  
  
  }

 }

}

elsif ($single_or_multiple_ch >='2' && $reference=='yes') {

my $count=1;

for my $i (1..$trtgroup_nb) {

  my $subjectcount=0;
  #initializes a counter in order to provide unique identifier to study subject
  
  for  my $j (1..$subject_nb) {
  
  $subjectcount=($i-1)*$subject_nb+$j;
  #computes the actual number of subjects by summing up the current rank of a subject in a given study group to the number of subjects already created
    
    for  my $k (0..$#target_tissue) {
    
    my $tissue=$target_tissue[$k];
    $k++;
    
    $record=join('  ', 
    "file-".$count.".txt",
    $array, 
    "group-".$i."-subject-".$subjectcount,
    $organism,
    $tissue,
    $cellmodel,
    "group-".$i."-subject-".$subjectcount.".sample-".$k,
    "P-DIET-1",
    "group-".$i."-subject-".$subjectcount.".sample-".$k."-extract",
    "P-EXTR-1",
    "group-".$i."-subject-".$subjectcount.".sample-".$k."-extract-"."le",
    "P-LABL-1",
    "Cy3",
    "hybridization-".$count,
    "P-HYBR-1",
    "scan-".$count,
    "P-SCAN-1",
    "trt_group".$i);
    
    #print "record: $record\n";

    push(@records,$record);
    
    
    $record=join('  ', 
    "file-".$count.".txt",
    $array, 
    "reference",
    $organism,
    "multi-tissue",
    "multi-cell type",
    "reference",
    "P-DIET-1",
    "reference",
    "P-EXTR-1",
    "reference",
    "P-LABL-1",
    "Cy5",
    "hybridization-".$count,
    "P-HYBR-1",
    "scan-".$count,
    "P-SCAN-1",
    "reference");
    push(@records,$record);
    
    $count++;

    
    }
  
  
  }



 }
 
 #TO IMPLEMENT: if reference=no, same stuff but assume matching sample at control level.
 
 
 

}



for my $l (0..$#records) {
print FILE "$records[$l]\n";
}

close FILE or die;

#---------------------------------
sub usage {
  print qq/
perl expert-agent.pl <OPTIONS>
-------------------------------------------------------------------------
 WARNINGS:

  ** Prior to running the script, REMEMBER TO PASS THE FOLLOWING COMMAND:
  limit datasize 1048000
-------------------------------------------------------------------------

 OPTIONS:

  --organism=s\t Name of organism under study (one only at the moment)
  --array=s\t Name of the array design used in study (one only at the moment)
  --trtgroup_nb=i\t The Number of study groups defined in the study (eg control, low dose, high dose would define 3 study groups)
  --subject_nb=i\t The number of subjects per study groups
  --target_tissue=s\t A comma separated list of organism parts (as in liver,abdominal adipose tissue, skeletal muscle)
  --dyeswap=s\t A semicolon separatedlist of assays whose technology is specified by csv between brackets.
  --single_or_multiple_channel=i\t An integer 1, 2 or 3
  --reference=s\t yes\no
-------------------------------------------------------------------------
 
 POST-PROCESSING:

   i.check and replace Protocol with relevant Accession Numbers
  ii.check\/add ExperimentalFactor categories
 iii.check\/create Person\/Organisation and AuditSecurity Package
  iv.if Final Transformed files are supplied, need to add those

-------------------------------------------------------------------------

/;
  exit(0);
}