From 9e8537997d84e78e6545a66a0d09c33e76c8b7cf Mon Sep 17 00:00:00 2001
From: Christoph Helma <helma@in-silico.ch>
Date: Fri, 30 Sep 2016 17:11:30 +0200
Subject: npo uri as source, spectral count unit f proteomics features

---
 lib/import.rb       | 31 +++++++++++++++++++++++++------
 lib/nanoparticle.rb | 18 ++++--------------
 2 files changed, 29 insertions(+), 20 deletions(-)

diff --git a/lib/import.rb b/lib/import.rb
index e187e3c..17894a9 100644
--- a/lib/import.rb
+++ b/lib/import.rb
@@ -62,24 +62,43 @@ module OpenTox
           np["bundles"].keys.each do |bundle_uri|
             nanoparticle.dataset_ids << datasets[bundle_uri].id
           end
+
           dataset = datasets[np["bundles"].keys.first]
           proteomics_features = {}
+          category = study["protocol"]["topcategory"]
+          source = study["protocol"]["category"]["term"]
+
           study["effects"].each do |effect|
+
             effect["result"]["textValue"] ?  klass = NominalFeature : klass = NumericFeature
             effect["conditions"].delete_if { |k, v| v.nil? }
+
             if study["protocol"]["category"]["title"].match(/Proteomics/) and effect["result"]["textValue"] and effect["result"]["textValue"].length > 50 # parse proteomics data
+
               JSON.parse(effect["result"]["textValue"]).each do |identifier, value| # time critical step
-                proteomics_features[identifier] ||= NumericFeature.find_or_create_by(:name => identifier, :category => "Proteomics")
+                proteomics_features[identifier] ||= NumericFeature.find_or_create_by(:name => identifier, :category => "Proteomics", :unit => "Spectral counts", :source => source)
                 nanoparticle.parse_ambit_value proteomics_features[identifier], value, dataset
               end
             else
               name = effect["endpoint"]
-              name = "log2(Net cell association)" if name == "Log2 transformed" # use a sensible name
+              unit = effect["result"]["unit"]
+              warnings = []
+              case name
+              when "Log2 transformed" # use a sensible name
+                name = "log2(Net cell association)"
+                warnings = ["Original name was 'Log2 transformed'"]
+                unit = "log2(mL/ug(Mg))"
+              when "Total protein (BCA assay)"
+                category = "P-CHEM"
+                warnings = ["Category changed from TOX to P-CHEM"]
+              end
               feature = klass.find_or_create_by(
-                :name => effect["endpoint"],
-                :unit => effect["result"]["unit"],
-                :category => study["protocol"]["topcategory"],
-                :conditions => effect["conditions"]
+                :name => name,
+                :unit => unit,
+                :category => category,
+                :conditions => effect["conditions"],
+                :source => study["protocol"]["category"]["term"],
+                :warnings => warnings
               )
               nanoparticle.parse_ambit_value feature, effect["result"], dataset
             end
diff --git a/lib/nanoparticle.rb b/lib/nanoparticle.rb
index d6261ee..b1a3835 100644
--- a/lib/nanoparticle.rb
+++ b/lib/nanoparticle.rb
@@ -5,7 +5,7 @@ module OpenTox
 
     field :core, type: Hash, default: {}
     field :coating, type: Array, default: []
-    field :proteomics, type: Hash, default: {}
+    #field :proteomics, type: Hash, default: {}
 
     attr_accessor :scaled_values
  
@@ -63,26 +63,16 @@ module OpenTox
           physchem_descriptors[feature.id.to_s] << value
           physchem_descriptors[feature.id.to_s].uniq!
         when "Proteomics"
-          #proteomics[feature.id.to_s] ||= []
-          #proteomics[feature.id.to_s] << value
-          #proteomics[feature.id.to_s].uniq!
           physchem_descriptors[feature.id.to_s] ||= []
           physchem_descriptors[feature.id.to_s] << value
           physchem_descriptors[feature.id.to_s].uniq!
         when "TOX"
-          if feature.name == "Total protein (BCA assay)"
-            physchem_descriptors[feature.id.to_s] ||= []
-            physchem_descriptors[feature.id.to_s] << value
-            physchem_descriptors[feature.id.to_s].uniq!
-          else
-            dataset.add self, feature, value
-          end
-          dataset.save
-          dataset_ids << dataset.id
-          dataset_ids.uniq!
+          dataset.add self, feature, value
         else
           warn "Unknown feature type '#{feature.category}'. Value '#{value}' not inserted."
         end
+        dataset_ids << dataset.id
+        dataset_ids.uniq!
       end
     end
 
-- 
cgit v1.2.3