#!/usr/bin/env ruby

header = ["ID","CID","SMILES","Canonical SMILES","Measured","lazar-MP2D","lazar-MP2D-high-confidence","lazar-PaDEL","lazar-PaDEL-high-confidence"]
tab = []
i = 0
File.read("pyrrolizidine-alkaloids/180920_PA_complete_SMILES.csv").each_line do |l|
  if i > 0
    id,cid,name,smi = l.chomp.split(";")
    tab << [id,cid,'"'+smi+'"']
  end
  i += 1
end
i = 0
File.read("pyrrolizidine-alkaloids/lazar/pa-mp2d-predictions.csv").each_line do |l|
  if i > 0
    id,cansmi,exp,mut,p0,p1,max_sim,nn = l.chomp.split(",")
    max_sim.to_f < 0.5? hc = "F" : hc = "T"
    hc = "" if mut.empty?
    tab[i-1] += ['"'+cansmi+'"',exp,mut,hc]
  end
  i += 1
end
i=0
File.read("pyrrolizidine-alkaloids/lazar/pa-padel-predictions.csv").each_line do |l|
  #if i > 0
    cansmi,exp,mut,p0,p1,max_sim,nn = l.chomp.split(",")
    max_sim.to_f < 0.5? hc = "F" : hc = "T"
    hc = "" if mut.empty?
    tab[i] += [mut,hc]
  #end
  i += 1
end
Dir["pyrrolizidine-alkaloids/R/PA.*.outcome.csv"].each do |r|
  header << "R-"+r.sub('pyrrolizidine-alkaloids/R/PA.','').sub('.outcome.csv','')
  i = 0
  File.read(r).each_line do |l|
    if i > 0
      items = l.chomp.split(";")
      items.shift
      if items.uniq.include? "1"
        tab[i-1] << 1
      elsif items.uniq.include? "0"
        tab[i-1] << 0
      end
    end
    i += 1
  end
end
Dir["pyrrolizidine-alkaloids/tensorflow/pred.*.v3-ext-Padel-2D.csv"].each do |r|
  header << "TF-"+r.sub('pyrrolizidine-alkaloids/tensorflow/pred.','').sub('.v3-ext-Padel-2D.csv','').sub("lr2","LR-scikit").sub("lr","LR-sgd").sub("rf","RF").sub("nn","NN")
  i = 0
  File.read(r).each_line do |l|
    if i > 0
      id,pred = l.chomp.split(",")
      pred.to_f > 0.5 ? tab[i-1] << 1 : tab[i-1] << 0
    end
    i += 1
  end
end
puts header.join(",")
puts tab.collect{|r| r.join(",")}.join("\n")