rbbt-text 1.3.0 → 1.3.5

data/test/rbbt/segment/test_named_entity.rb

@@ -22,12 +22,14 @@ class TestClass < Test::Unit::TestCase
     assert_equal "SCORE", a.score
   end
 
-  def __test_tsv
+  def test_tsv
     a = "test"
-    NamedEntity.setup a, 10, "TYPE", "CODE", "SCORE"
-    assert Segment.tsv([a]).fields.include? "code"
-    assert Segment.tsv([a], nil).fields.include? "code"
-    assert Segment.tsv([a], "literal").fields.include? "code"
+    NamedEntity.setup a, 10, "DocID", "TYPE", "CODE", "SCORE"
+    ppp Annotated.tsv([a,a])
+    assert Annotated.tsv([a]).fields.include? "code"
+    assert Annotated.tsv([a], nil).fields.include? "code"
+    assert Annotated.tsv([a], :all).fields.include? "code"
+    assert Annotated.tsv([a], :all).fields.include? "literal"
   end
 
   def __test_segment_brat

data/test/rbbt/segment/test_range_index.rb

@@ -9,8 +9,7 @@ class TestRangeIndex < Test::Unit::TestCase
     text = "This sentence mentions the TP53 gene and the CDK5R1 protein"
     Document.setup(text, "TEST", "test_doc1", nil)
 
-    corpus = {}
-    corpus.extend Document::Corpus
+    corpus = Document::Corpus.setup({})
 
     corpus.add_document(text)
 

data/test/rbbt/segment/test_transformed.rb

@@ -101,6 +101,35 @@ More recently, PPAR activators were shown to inhibit the activation of inflammat
     assert_equal "CDK5R1 protein", exp2
   end
 
+  def test_with_transform_sentences
+    a = "This first sentence mentions Bread. This sentence mentions the TP53 gene and the CDK5R1 protein"
+    original = a.dup
+
+    gene1 = "TP53"
+    gene1.extend NamedEntity
+    gene1.offset = a.index gene1
+
+    gene2 = "CDK5R1"
+    gene2.extend NamedEntity
+    gene2.offset = a.index gene2
+
+    bread = "Bread"
+    bread.extend NamedEntity
+    bread.offset = a.index bread
+
+    sentences = Segment.align(a, a.split(". "))
+
+    Transformed.with_transform(sentences[1], [gene1, gene2, bread], "GN") do
+      assert sentences[1].include?("GN gene and the GN protein")
+    end
+
+    Transformed.with_transform(sentences[0], [gene1, gene2, bread], "BR") do
+      assert sentences[0].include?("first sentence mentions BR")
+    end
+
+
+  end
+
   def test_html
     a = "This sentence mentions the TP53 gene and the CDK5R1 protein"
 
@@ -115,7 +144,7 @@ More recently, PPAR activators were shown to inhibit the activation of inflammat
     gene2.entity_type = "Protein"
 
     Transformed.with_transform(a, [gene1,gene2], Proc.new{|e| e.html}) do 
-      assert_equal "This sentence mentions the <span class='Entity' attr-entity-type='Gene'>TP53</span> gene and the <span class='Entity' attr-entity-type='Protein'>CDK5R1</span> protein", a
+      assert_equal "This sentence mentions the <span class='Entity' attr-entity-type='Gene' title='Gene'>TP53</span> gene and the <span class='Entity' attr-entity-type='Protein' title='Protein'>CDK5R1</span> protein", a
     end
   end
 
@@ -136,7 +165,7 @@ More recently, PPAR activators were shown to inhibit the activation of inflammat
     gene2.entity_type = "Protein"
 
     Transformed.with_transform(a, [gene1,gene2], Proc.new{|e| e.html}) do 
-      assert_equal "This sentence mentions the <span class='Entity' attr-entity-type='Gene'>TP53</span> gene and the <span class='Entity' attr-entity-type='Protein'>CDK5R1</span> protein", a
+      assert_equal "This sentence mentions the <span class='Entity' attr-entity-type='Gene' title='Gene'>TP53</span> gene and the <span class='Entity' attr-entity-type='Protein' title='Protein'>CDK5R1</span> protein", a
     end
   end
 
@@ -156,9 +185,9 @@ More recently, PPAR activators were shown to inhibit the activation of inflammat
     assert_equal [gene1], Segment.overlaps(Segment.sort([gene1,gene2]))
 
     Transformed.with_transform(a, [gene1], Proc.new{|e| e.html}) do 
-      assert_equal "This sentence mentions the <span class='Entity' attr-entity-type='Gene'>TP53</span> gene and the CDK5R1 protein", a
+      assert_equal "This sentence mentions the <span class='Entity' attr-entity-type='Gene' title='Gene'>TP53</span> gene and the CDK5R1 protein", a
       Transformed.with_transform(a, [gene2], Proc.new{|e| e.html}) do 
-        assert_equal "This sentence mentions the <span class='Entity' attr-entity-type='Expanded Gene'><span class='Entity' attr-entity-type='Gene'>TP53</span> gene</span> and the CDK5R1 protein", a
+        assert_equal "This sentence mentions the <span class='Entity' attr-entity-type='Expanded Gene' title='Expanded Gene'><span class='Entity' attr-entity-type='Gene' title='Gene'>TP53</span> gene</span> and the CDK5R1 protein", a
       end
     end
   end

data/test/rbbt/test_segment.rb

@@ -17,8 +17,7 @@ class TestSegment < Test::Unit::TestCase
     text = "This is a document"
     Document.setup(text, "TEST", "test_doc1", nil)
 
-    corpus = {}
-    corpus.extend Document::Corpus
+    corpus = Document::Corpus.setup({})
 
     corpus.add_document(text)
 
@@ -41,8 +40,7 @@ class TestSegment < Test::Unit::TestCase
     text = "This sentence mentions the TP53 gene and the CDK5R1 protein"
     Document.setup(text, "TEST", "test_doc1", nil)
 
-    corpus = {}
-    corpus.extend Document::Corpus
+    corpus = Document::Corpus.setup({})
 
     corpus.add_document(text)
 
@@ -65,8 +63,7 @@ class TestSegment < Test::Unit::TestCase
     text = "This sentence mentions the TP53 gene and the CDK5R1 protein"
     Document.setup(text, "TEST", "test_doc1", nil)
 
-    corpus = {}
-    corpus.extend Document::Corpus
+    corpus = Document::Corpus.setup({})
 
     corpus.add_document(text)
 
@@ -94,8 +91,7 @@ class TestSegment < Test::Unit::TestCase
     text = "This sentence mentions the TP53 gene and the CDK5R1 protein"
     Document.setup(text, "TEST", "test_doc1", nil)
 
-    corpus = {}
-    corpus.extend Document::Corpus
+    corpus = Document::Corpus.setup({})
 
     corpus.add_document(text)
 
@@ -142,8 +138,7 @@ Atypical teratoid/rhabdoid tumors (AT/RTs) are highly aggressive brain tumors of
     text = "This sentence mentions the TP53 gene and the CDK5R1 protein"
     Document.setup(text, "TEST", "test_doc1", nil)
 
-    corpus = {}
-    corpus.extend Document::Corpus
+    corpus = Document::Corpus.setup({})
 
     corpus.add_document(text)
 

data/test/test_spaCy.rb

@@ -0,0 +1,144 @@
+require File.join(File.expand_path(File.dirname(__FILE__)), '', 'test_helper.rb')
+require 'rbbt/nlp/spaCy'
+require 'rbbt/document/corpus'
+
+class TestSpaCy < Test::Unit::TestCase
+  def test_tokens
+    text = "I tell a story"
+
+    tokens = SpaCy.tokens(text)
+
+    assert_equal 4, tokens.length
+    assert_equal "tell", tokens[1].to_s
+  end
+
+  def test_chunks
+    text = "Miguel Vazquez tell a good story"
+
+    tokens = SpaCy.chunks(text)
+
+    assert_equal 2, tokens.length
+    assert_equal "Miguel Vazquez", tokens[0].to_s
+  end
+
+
+  def test_segments
+    text = "I tell a story. It's a very good story."
+
+    corpus = Document::Corpus.setup({})
+
+    Document.setup(text, "TEST", "test_doc1", "simple_sentence")
+
+    corpus.add_document text
+    text.corpus = corpus
+
+    segments = SpaCy.segments(text)
+
+    segments.each do |segment|
+      assert_equal segment, segment.segid.tap{|e| e.corpus = corpus}.segment
+    end
+  end
+
+  def test_chunk_segments
+    text = "I tell a story. It's a very good story."
+
+    corpus = Document::Corpus.setup({})
+
+    Document.setup(text, "TEST", "test_doc1", "simple_sentence")
+
+    corpus.add_document text
+    text.corpus = corpus
+
+    segments = SpaCy.chunk_segments(text)
+
+    segments.each do |segment|
+      assert_equal segment, segment.segid.tap{|e| e.corpus = corpus}.segment
+    end
+  end
+
+  def test_dep_graph
+    text = "Meanwhile, TF antisense treatment activated the human ASBT promoter 5-fold and not only abrogated interleukin-1beta-mediated repression but led to a paradoxical increase in TG promoter activity"
+    graph = SpaCy.dep_graph(text, true)
+
+    tokens = SpaCy.segments(text)
+    index = Segment.index tokens
+    tf_s = tokens.select{|t| t == "TF" }.first
+    tg_s = tokens.select{|t| t == "TG" }.first
+
+    require 'rbbt/network/paths'
+
+    path = Paths.dijkstra(graph, tf_s.segid, [tg_s.segid])
+    path_tokens = path.collect do |segid|
+      range = Range.new(*segid.split(":").last.split("..").map(&:to_i))
+      text[range]
+    end
+
+    assert path_tokens.include? 'increase'
+
+  end
+
+  def test_chunk_dep_graph
+    text = "Meanwhile, TF antisense treatment activated the human ASBT promoter 5-fold and not only abrogated interleukin-1beta-mediated repression but led to a paradoxical increase in TG promoter activity"
+    graph = SpaCy.chunk_dep_graph(text, true)
+
+    tokens = SpaCy.chunk_segments(text)
+    index = Segment.index tokens
+    tf_s = tokens.select{|t| t.include? "TF" }.first
+    tg_s = tokens.select{|t| t.include? "TG" }.first
+
+
+    require 'rbbt/network/paths'
+
+    path = Paths.dijkstra(graph, tf_s.segid, [tg_s.segid])
+    path_tokens = path.collect do |segid|
+      range = Range.new(*segid.split(":").last.split("..").map(&:to_i))
+      text[range]
+    end
+
+    assert path_tokens.include? 'increase'
+  end
+
+  def test_paths
+    text = "Meanwhile, TF antisense treatment activated the human ASBT promoter 5-fold and not only abrogated interleukin-1beta-mediated repression but led to a paradoxical increase in TG promoter activity"
+    path = SpaCy.paths(text, Segment.setup("TF", :offset => text.index("TF")), Segment.setup("TG",:offset =>  text.index("TG")))
+
+
+    path_tokens = path.collect do |segid|
+      range = Range.new(*segid.split(":").last.split("..").map(&:to_i))
+      text[range]
+    end
+
+    ppp text
+    iii path_tokens
+
+    assert path_tokens.include? 'increase'
+  end
+
+  def test_paths2
+    text = "Deletion and domain swap experiments identified small, discreet positive and negative elements in A-Myb and TF that were required for the regulation of specific genes, such as DHRS2, TG, and mim-1"
+    path = SpaCy.paths(text, Segment.setup("TF", :offset => text.index("TF")), Segment.setup("TG",:offset =>  text.index("TG")))
+
+
+    path_tokens = path.collect do |segid|
+      range = Range.new(*segid.split(":").last.split("..").map(&:to_i))
+      text[range]
+    end
+    
+    iii path_tokens
+
+
+    assert path_tokens.include? 'regulation'
+  end
+
+  def test_paths3
+    text = "Therefore, we speculate that PEA3 factors may contribute to the up-regulation of COX-2 expression resulting from both APC mutation and Wnt1 expression"
+    path = SpaCy.paths(text, *Segment.align(text,["PEA3", "Wnt1"]))
+
+    path_tokens = path.collect do |segid|
+      range = Range.new(*segid.split(":").last.split("..").map(&:to_i))
+      text[range]
+    end
+
+  end
+end
+

metadata

@@ -1,14 +1,14 @@
 --- !ruby/object:Gem::Specification
 name: rbbt-text
 version: !ruby/object:Gem::Version
-  version: 1.3.0
+  version: 1.3.5
 platform: ruby
 authors:
 - Miguel Vazquez
 autorequire: 
 bindir: bin
 cert_chain: []
-date: 2020-04-18 00:00:00.000000000 Z
+date: 2021-06-17 00:00:00.000000000 Z
 dependencies:
 - !ruby/object:Gem::Dependency
   name: rbbt-util
@@ -95,6 +95,7 @@ files:
 - lib/rbbt/ner/oscar4.rb
 - lib/rbbt/ner/patterns.rb
 - lib/rbbt/ner/regexpNER.rb
+- lib/rbbt/ner/rner.rb
 - lib/rbbt/ner/rnorm.rb
 - lib/rbbt/ner/rnorm/cue_index.rb
 - lib/rbbt/ner/rnorm/tokens.rb
@@ -102,12 +103,15 @@ files:
 - lib/rbbt/nlp/genia/sentence_splitter.rb
 - lib/rbbt/nlp/nlp.rb
 - lib/rbbt/nlp/open_nlp/sentence_splitter.rb
+- lib/rbbt/nlp/spaCy.rb
+- lib/rbbt/relationship.rb
 - lib/rbbt/segment.rb
 - lib/rbbt/segment/annotation.rb
 - lib/rbbt/segment/encoding.rb
 - lib/rbbt/segment/named_entity.rb
 - lib/rbbt/segment/overlaps.rb
 - lib/rbbt/segment/range_index.rb
+- lib/rbbt/segment/relationship.rb
 - lib/rbbt/segment/segmented.rb
 - lib/rbbt/segment/token.rb
 - lib/rbbt/segment/transformed.rb
@@ -124,6 +128,7 @@ files:
 - share/install/software/OpenNLP
 - share/install/software/StanfordParser
 - share/patterns/drug_induce_disease
+- share/rner/config.rb
 - share/rnorm/cue_default
 - share/rnorm/tokens_default
 - share/wordlists/stopwords
@@ -146,6 +151,7 @@ files:
 - test/rbbt/ner/test_oscar4.rb
 - test/rbbt/ner/test_patterns.rb
 - test/rbbt/ner/test_regexpNER.rb
+- test/rbbt/ner/test_rner.rb
 - test/rbbt/ner/test_rnorm.rb
 - test/rbbt/ner/test_token_trieNER.rb
 - test/rbbt/nlp/genia/test_sentence_splitter.rb
@@ -161,6 +167,7 @@ files:
 - test/rbbt/test_document.rb
 - test/rbbt/test_segment.rb
 - test/test_helper.rb
+- test/test_spaCy.rb
 homepage: http://github.com/mikisvaz/rbbt-util
 licenses: []
 metadata: {}
@@ -179,7 +186,7 @@ required_rubygems_version: !ruby/object:Gem::Requirement
     - !ruby/object:Gem::Version
       version: '0'
 requirements: []
-rubygems_version: 3.0.6
+rubygems_version: 3.1.4
 signing_key: 
 specification_version: 4
 summary: Text mining tools for the Ruby Bioinformatics Toolkit (rbbt)
@@ -207,6 +214,7 @@ test_files:
 - test/rbbt/ner/test_banner.rb
 - test/rbbt/ner/test_token_trieNER.rb
 - test/rbbt/ner/test_finder.rb
+- test/rbbt/ner/test_rner.rb
 - test/rbbt/ner/test_linnaeus.rb
 - test/rbbt/ner/test_oscar4.rb
 - test/rbbt/test_segment.rb
@@ -217,4 +225,5 @@ test_files:
 - test/rbbt/segment/test_encoding.rb
 - test/rbbt/segment/test_range_index.rb
 - test/rbbt/segment/test_corpus.rb
+- test/test_spaCy.rb
 - test/test_helper.rb