1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
|
require_relative '../../lazar/lib/lazar'
include OpenTox
dirpath = File.join(File.dirname(__FILE__),"..","regression")
old = CSV.read File.join(dirpath,"LOAEL_mg_corrected_smiles_mmol.csv")
old.shift
new = CSV.read File.join(dirpath,"swissRat_chron_LOAEL_mmol.csv")
new.shift
p old.size
p new.size
# canonical smiles
old.collect!{|r| [Compound.from_smiles(r.first).smiles, r.last]}
new.collect!{|r| [Compound.from_smiles(r.first).smiles, r.last]}
old_compounds = old.collect{|r| r.first}.uniq
new_compounds = new.collect{|r| r.first}.uniq
p old_compounds.size
p new_compounds.size
common_compounds = (old_compounds & new_compounds).uniq
p common_compounds.size
common = []
# TODO: canonical smiles??
common_compounds.each do |smi|
old_rows = old.select{|r| r.first == smi}
new_rows = new.select{|r| r.first == smi}
common += old_rows + new_rows
old -= old_rows
new -= new_rows
end
header = ["SMILES","LOAEL"]
p old.size
p new.size
p common.size
{
"mazzatorta-loael-training.csv" => old.uniq,
"swiss-loael-training.csv" => new.uniq,
"combined-training.csv" => (old+new).uniq,
"common-test.csv" => common.uniq,
}.each do |file,data|
CSV.open(File.join(dirpath,file),"w+") do |csv|
csv << header
data.each{|row| csv << row}
end
end
|
