biointerchange 0.1.2 → 0.1.3
Sign up to get free protection for your applications and to get access to all the features.
- data/README.md +49 -4
- data/VERSION +1 -1
- data/examples/chromosome_BF.gff +1701 -0
- data/examples/estd176_Banerjee_et_al_2011.2012-11-29.NCBI36.gvf +4326 -0
- data/examples/pubannotation.10096561.json +1 -0
- data/examples/{pubannotation.json → pubannotation.10096561.json.old} +0 -0
- data/examples/pubannotation.2626671.json +1 -0
- data/lib/biointerchange/core.rb +58 -16
- data/lib/biointerchange/genomics/gff3_feature.rb +1 -0
- data/lib/biointerchange/genomics/gff3_feature_set.rb +31 -1
- data/lib/biointerchange/genomics/gff3_pragmas.rb +35 -0
- data/lib/biointerchange/genomics/gff3_rdf_ntriples.rb +60 -23
- data/lib/biointerchange/genomics/gff3_reader.rb +74 -40
- data/lib/biointerchange/genomics/gvf_feature.rb +24 -0
- data/lib/biointerchange/genomics/gvf_feature_set.rb +14 -0
- data/lib/biointerchange/genomics/gvf_pragmas.rb +6 -0
- data/lib/biointerchange/genomics/gvf_reader.rb +37 -0
- data/lib/biointerchange/gff3o.rb +1 -1
- data/lib/biointerchange/gvf1o.rb +145 -17
- data/lib/biointerchange/textmining/content.rb +1 -0
- data/lib/biointerchange/textmining/content_connection.rb +74 -0
- data/lib/biointerchange/textmining/document.rb +3 -1
- data/lib/biointerchange/textmining/pubannos_json_reader.rb +87 -9
- data/lib/biointerchange/textmining/text_mining_rdf_ntriples.rb +58 -2
- data/spec/gff3_rdfwriter_spec.rb +9 -1
- data/spec/gvf_rdfwriter_spec.rb +81 -0
- data/spec/text_mining_pubannos_json_reader_spec.rb +82 -10
- data/spec/text_mining_rdfwriter_spec.rb +11 -0
- data/web/api.html +30 -23
- metadata +156 -138
@@ -21,7 +21,9 @@ class Document
|
|
21
21
|
#
|
22
22
|
# +content+:: content of type +BioInterchange::TextMining::Content+ that should be added to the document
|
23
23
|
def add(content)
|
24
|
-
|
24
|
+
if ( (! content.kind_of?(BioInterchange::TextMining::Content)) && (! content.kind_of?(BioInterchange::TextMining::ContentConnection)) )
|
25
|
+
raise BioInterchange::Exceptions::ImplementationModelError, 'Content has to be of kind BioInterchange::TextMining::Content or kind BioInterchange::TextMining::ContentConnection'
|
26
|
+
end
|
25
27
|
@content << content
|
26
28
|
end
|
27
29
|
|
@@ -27,7 +27,7 @@ private
|
|
27
27
|
raise BioInterchange::Exceptions::InputFormatError, 'Error parsing the JSON input file: #{result["Error"]}'
|
28
28
|
end
|
29
29
|
|
30
|
-
|
30
|
+
|
31
31
|
text = result['text']
|
32
32
|
#doc_uri = "http://pubannotation.dbcls.jp/pmdocs/" + result['pmid'].to_s
|
33
33
|
doc_uri = result['docurl']
|
@@ -39,18 +39,25 @@ private
|
|
39
39
|
|
40
40
|
#so our document requires content of type document or abstract
|
41
41
|
#should it hold the content string?
|
42
|
-
|
42
|
+
|
43
|
+
#hash to remember annotation in case they are needed for building upon based on ids later
|
44
|
+
contents = {}
|
45
|
+
|
43
46
|
if result['catanns']
|
44
47
|
result['catanns'].each do |annot|
|
45
|
-
start_offset =
|
46
|
-
end_offset =
|
48
|
+
start_offset = 0
|
49
|
+
end_offset = 0
|
50
|
+
if annot['span']
|
51
|
+
start_offset = annot['span']['begin']
|
52
|
+
end_offset = annot['span']['end']
|
53
|
+
elsif annot['begin'] and annot['end']
|
54
|
+
start_offset = annot['begin']
|
55
|
+
end_offset = annot['end']
|
56
|
+
end
|
47
57
|
length = end_offset - start_offset
|
48
|
-
|
49
|
-
updated_time = annot['updated_at']
|
58
|
+
|
50
59
|
category = annot['category']
|
51
|
-
|
52
|
-
#doc_id = annot['doc_id']
|
53
|
-
#id = annot['id']
|
60
|
+
id = annot['id']
|
54
61
|
|
55
62
|
entity = text.slice(start_offset..end_offset)
|
56
63
|
|
@@ -58,11 +65,82 @@ private
|
|
58
65
|
con = Content.new(start_offset, length, Content::PHRASE, @process)
|
59
66
|
con.setContext(doc)
|
60
67
|
doc.add(con)
|
68
|
+
|
69
|
+
contents[id] = con
|
61
70
|
|
62
71
|
#set process.date = updated_time?
|
63
72
|
end
|
64
73
|
end
|
65
74
|
|
75
|
+
if result['insanns']
|
76
|
+
result['insanns'].each do |annot|
|
77
|
+
|
78
|
+
#unsure what to do about this (con1), 'E1' is the ID of something not created yet.
|
79
|
+
#it is perhaps a case of making a new content, but with what params...?
|
80
|
+
#need to conform what this is refering to with JDK
|
81
|
+
con1 = nil
|
82
|
+
con2 = contents[annot['object']]
|
83
|
+
|
84
|
+
#get annotation type
|
85
|
+
type = ContentConnection::UNSPECIFIED
|
86
|
+
case annot['type']
|
87
|
+
when 'subClassOf'
|
88
|
+
type = ContentConnection::SUBCLASS
|
89
|
+
end
|
90
|
+
connection = ContentConnection.new(con1, con2, type, @process)
|
91
|
+
connection.setContext(doc)
|
92
|
+
doc.add(connection)
|
93
|
+
|
94
|
+
contents[annot['id']] = connection
|
95
|
+
|
96
|
+
end
|
97
|
+
end
|
98
|
+
|
99
|
+
if result['relanns']
|
100
|
+
result['relanns'].each do |annot|
|
101
|
+
con1 = contents[annot['subject']]
|
102
|
+
con2 = contents[annot['object']]
|
103
|
+
|
104
|
+
#get annotation type
|
105
|
+
type = ContentConnection::UNSPECIFIED
|
106
|
+
case annot['type']
|
107
|
+
when 'equivalentTo'
|
108
|
+
type = ContentConnection::EQUIVALENCE
|
109
|
+
when 'themeOf'
|
110
|
+
type = ContentConnection::THEME
|
111
|
+
end
|
112
|
+
connection = ContentConnection.new(con1, con2, type, @process)
|
113
|
+
connection.setContext(doc)
|
114
|
+
doc.add(connection)
|
115
|
+
|
116
|
+
contents[annot['id']] = connection
|
117
|
+
|
118
|
+
end
|
119
|
+
end
|
120
|
+
|
121
|
+
if result['modanns']
|
122
|
+
result['modanns'].each do |annot|
|
123
|
+
|
124
|
+
#in this case, it is a modification of an already existing content object (speculation/negation).
|
125
|
+
con = contents[annot['object']]
|
126
|
+
|
127
|
+
#get annotation type
|
128
|
+
type = ContentConnection::UNSPECIFIED
|
129
|
+
case annot['type']
|
130
|
+
when 'Speculation'
|
131
|
+
type = ContentConnection::SPECULATION
|
132
|
+
when 'Negation'
|
133
|
+
type = ContentConnection::NEGATION
|
134
|
+
end
|
135
|
+
connection = ContentConnection.new(con, nil, type, @process)
|
136
|
+
connection.setContext(doc)
|
137
|
+
doc.add(connection)
|
138
|
+
|
139
|
+
contents[annot['id']] = connection
|
140
|
+
|
141
|
+
end
|
142
|
+
end
|
143
|
+
|
66
144
|
doc
|
67
145
|
end
|
68
146
|
|
@@ -10,7 +10,7 @@ class RDFWriter < BioInterchange::Writer
|
|
10
10
|
#
|
11
11
|
# +ostream+:: instance of an IO class or derivative that is used for RDF serialization
|
12
12
|
def initialize(ostream)
|
13
|
-
raise BioInterchange::Exceptions::ImplementationWriterError, 'The output stream is not an instance of IO or its subclasses.' unless ostream.kind_of?(IO)
|
13
|
+
raise BioInterchange::Exceptions::ImplementationWriterError, 'The output stream is not an instance of IO or its subclasses.' unless ostream.kind_of?(IO) || ostream.kind_of?(StringIO)
|
14
14
|
@ostream = ostream
|
15
15
|
end
|
16
16
|
|
@@ -68,6 +68,22 @@ private
|
|
68
68
|
end
|
69
69
|
end
|
70
70
|
|
71
|
+
# Generates an URI for a given contentconnection and its contents.
|
72
|
+
#
|
73
|
+
# +contentcon+:: content connection instance
|
74
|
+
# +kind+:: kind of the URI that should be generated, for example, whether the URI should represent the name, date, etc.
|
75
|
+
def content_connection_uri(contentcon, kind)
|
76
|
+
base_uri = 'biointerchange://textmining/content_connection'
|
77
|
+
case kind
|
78
|
+
when :start
|
79
|
+
RDF::URI.new("#{base_uri}/start/#{content.uri.sub(/^.*?:\/\//, '')}")
|
80
|
+
when :stop
|
81
|
+
RDF::URI.new("#{base_uri}/stop/#{content.uri.sub(/^.*?:\/\//, '')}")
|
82
|
+
else
|
83
|
+
raise BioInterchange::Exceptions::ImplementationWriterError, "There is no implementation for serializing a content as #{kind}."
|
84
|
+
end
|
85
|
+
end
|
86
|
+
|
71
87
|
# Serializes RDF for a textual document representation using the Semanticsciene Integrated Ontology
|
72
88
|
# (http://code.google.com/p/semanticscience/wiki/SIO).
|
73
89
|
#
|
@@ -77,7 +93,13 @@ private
|
|
77
93
|
document_uri = RDF::URI.new(model.uri)
|
78
94
|
graph.insert(RDF::Statement.new(document_uri, RDF.type, BioInterchange::SIO.document))
|
79
95
|
model.contents.each { |content|
|
80
|
-
|
96
|
+
if content.kind_of?(BioInterchange::TextMining::Content)
|
97
|
+
serialize_content(graph, document_uri, content)
|
98
|
+
elsif content.kind_of?(BioInterchange::TextMining::ContentConnection)
|
99
|
+
serialize_contentconnection(graph, document_uri, content)
|
100
|
+
else
|
101
|
+
raise BioInterchange::Exceptions::ImplementationWriterError, "Can only serialize Content and ContentConnection from a Document."
|
102
|
+
end
|
81
103
|
}
|
82
104
|
RDF::NTriples::Writer.dump(graph, @ostream)
|
83
105
|
end
|
@@ -128,6 +150,40 @@ private
|
|
128
150
|
|
129
151
|
end
|
130
152
|
|
153
|
+
# Serializes a ContentConnection object for a given document URI.
|
154
|
+
#
|
155
|
+
# +graph+:: RDF graph to which content is added
|
156
|
+
# +document_uri+:: the document URI to which the added content belongs to
|
157
|
+
# +content+:: an instance that describes the content
|
158
|
+
def serialize_contentconnection(graph, document_uri, contentcon)
|
159
|
+
contentcon_uri = RDF::URI.new(contentcon.uri)
|
160
|
+
graph.insert(RDF::Statement.new(document_uri, BioInterchange::SIO.has_attribute, contentcon_uri))
|
161
|
+
serialize_process(graph, document_uri, contentcon_uri, contentcon.process) if contentcon.process
|
162
|
+
|
163
|
+
|
164
|
+
#TODO these sio tags need confirming - there are here as a initial proof of concept
|
165
|
+
#next issue, some of these are relations and some are labels, need to separate out which
|
166
|
+
#I seem to recall that the only relationship types that should be used are "has_attribute" and "RDF::type", in which case these need adjusting for that.
|
167
|
+
#I presume this'd mean making a "has_attribute" link between the content1 and the contentconnection relationship in some way.
|
168
|
+
case contentcon.type
|
169
|
+
when ContentConnection::UNSPECIFIED
|
170
|
+
graph.insert(RDF::Statement.new(contentcon.content1.uri, BioInterchange::SIO.has_attribute, BioInterchange::SIO.language_entity))
|
171
|
+
when ContentConnection::EQUIVALENCE
|
172
|
+
graph.insert(RDF::Statement.new(contentcon.content1.uri, BioInterchange::SIO.is_equal_to, contentcon.content2.uri))
|
173
|
+
when ContentConnection::SUBCLASS
|
174
|
+
#TODO this class needs more information, the relationship is between a contentcon.content, and 'something'... I've yet to work out what
|
175
|
+
graph.insert(RDF::Statement.new(contentcon.content2.uri, BioInterchange::SIO.has_attribute, BioInterchange::SIO.in_relation_to))
|
176
|
+
when ContentConnection::THEME
|
177
|
+
#TODO there are other more specific options for this that need investigating as options.
|
178
|
+
graph.insert(RDF::Statement.new(contentcon.content1.uri, BioInterchange::SIO.has_target, contentcon.content2.uri))
|
179
|
+
when ContentConnection::SPECULATION
|
180
|
+
graph.insert(RDF::Statement.new(contentcon.content1.uri, BioInterchange::SIO.has_attribute, BioInterchange::SIO.speculation))
|
181
|
+
when ContentConnection::NEGATION
|
182
|
+
graph.insert(RDF::Statement.new(contentcon.content1.uri, BioInterchange::SIO.denotes, BioInterchange::SIO.negative_regulation))
|
183
|
+
end
|
184
|
+
|
185
|
+
end
|
186
|
+
|
131
187
|
# Serializes a process object for a specific document uri
|
132
188
|
#
|
133
189
|
#
|
data/spec/gff3_rdfwriter_spec.rb
CHANGED
@@ -66,7 +66,15 @@ describe BioInterchange::Genomics::RDFWriter do
|
|
66
66
|
set.add(feature)
|
67
67
|
BioInterchange::Genomics::RDFWriter.new(ostream).serialize(set)
|
68
68
|
ostream.close
|
69
|
-
istream.read.lines
|
69
|
+
lines = istream.read.lines
|
70
|
+
feature_no = 0
|
71
|
+
lines.each { |line|
|
72
|
+
subject, predicate, object = line.chomp.split(/\s/, 3)
|
73
|
+
object.sub!(/\s+\.$/, '')
|
74
|
+
feature_no += 1 if predicate == "<#{RDF.type}>" and object == "<#{BioInterchange::GFF3O.Feature}>"
|
75
|
+
}
|
76
|
+
lines.count.should be == 43
|
77
|
+
feature_no.should be == 3
|
70
78
|
end
|
71
79
|
end
|
72
80
|
end
|
@@ -0,0 +1,81 @@
|
|
1
|
+
|
2
|
+
require 'rubygems'
|
3
|
+
require 'rspec'
|
4
|
+
|
5
|
+
# Turn off verbose reporting here, since class definitions may be loaded multiple
|
6
|
+
# times here. That reports that constants have been already been initialized, which
|
7
|
+
# is true, but they are only "re-initialized" with the very same values.
|
8
|
+
v, $VERBOSE = $VERBOSE, nil
|
9
|
+
load 'lib/biointerchange/core.rb'
|
10
|
+
load 'lib/biointerchange/gvf1o.rb'
|
11
|
+
load 'lib/biointerchange/sofa.rb'
|
12
|
+
load 'lib/biointerchange/reader.rb'
|
13
|
+
load 'lib/biointerchange/writer.rb'
|
14
|
+
load 'lib/biointerchange/genomics/gvf_feature_set.rb'
|
15
|
+
load 'lib/biointerchange/genomics/gvf_feature.rb'
|
16
|
+
# The GVF implementation extends the GFF3 implementation, so load those classes too:
|
17
|
+
load 'lib/biointerchange/genomics/gff3_rdf_ntriples.rb'
|
18
|
+
load 'lib/biointerchange/genomics/gff3_feature_set.rb'
|
19
|
+
load 'lib/biointerchange/genomics/gff3_feature.rb'
|
20
|
+
$VERBOSE = v
|
21
|
+
|
22
|
+
describe BioInterchange::Genomics::RDFWriter do
|
23
|
+
describe 'serialization of GVF models' do
|
24
|
+
it 'empty document' do
|
25
|
+
istream, ostream = IO.pipe
|
26
|
+
BioInterchange::Genomics::RDFWriter.new(ostream).serialize(BioInterchange::Genomics::GVFFeatureSet.new())
|
27
|
+
ostream.close
|
28
|
+
istream.read.lines.count.should eq(1)
|
29
|
+
end
|
30
|
+
|
31
|
+
it 'model with three features' do
|
32
|
+
istream, ostream = IO.pipe
|
33
|
+
set = BioInterchange::Genomics::GVFFeatureSet.new()
|
34
|
+
feature = BioInterchange::Genomics::GVFFeature.new(
|
35
|
+
'GRCh37.1',
|
36
|
+
'NCBI',
|
37
|
+
BioInterchange::SOFA.CDS,
|
38
|
+
32890598,
|
39
|
+
32890664,
|
40
|
+
0.1,
|
41
|
+
BioInterchange::Genomics::GFF3Feature::POSITIVE,
|
42
|
+
{ 'ID' => [ 'BRCA2' ], 'annotation' => [ 'manual' ] }
|
43
|
+
)
|
44
|
+
set.add(feature)
|
45
|
+
feature = BioInterchange::Genomics::GVFFeature.new(
|
46
|
+
'GRCh37.1',
|
47
|
+
'NCBI',
|
48
|
+
BioInterchange::SOFA.modified_base,
|
49
|
+
32890599,
|
50
|
+
32890599,
|
51
|
+
0.8,
|
52
|
+
BioInterchange::Genomics::GFF3Feature::POSITIVE,
|
53
|
+
{ 'ID' => [ 'aModifiedBase' ], 'Parent' => [ 'BRCA2' ] }
|
54
|
+
)
|
55
|
+
set.add(feature)
|
56
|
+
feature = BioInterchange::Genomics::GVFFeature.new(
|
57
|
+
'GRCh37.1',
|
58
|
+
'NCBI',
|
59
|
+
BioInterchange::SOFA.modified_base,
|
60
|
+
32890599,
|
61
|
+
32890599,
|
62
|
+
0.8,
|
63
|
+
BioInterchange::Genomics::GFF3Feature::POSITIVE,
|
64
|
+
{ 'Parent' => [ 'BRCA2', 'aModifiedBase' ] }
|
65
|
+
)
|
66
|
+
set.add(feature)
|
67
|
+
BioInterchange::Genomics::RDFWriter.new(ostream).serialize(set)
|
68
|
+
ostream.close
|
69
|
+
lines = istream.read.lines
|
70
|
+
feature_no = 0
|
71
|
+
lines.each { |line|
|
72
|
+
subject, predicate, object = line.chomp.split(/\s/, 3)
|
73
|
+
object.sub!(/\s+\.$/, '')
|
74
|
+
feature_no += 1 if predicate == "<#{RDF.type}>" and object == "<#{BioInterchange::GVF1O.Feature}>"
|
75
|
+
}
|
76
|
+
lines.count.should be == 43
|
77
|
+
feature_no.should be == 3
|
78
|
+
end
|
79
|
+
end
|
80
|
+
end
|
81
|
+
|
@@ -12,6 +12,7 @@ load 'lib/biointerchange/textmining/text_mining_reader.rb'
|
|
12
12
|
load 'lib/biointerchange/textmining/pubannos_json_reader.rb'
|
13
13
|
load 'lib/biointerchange/textmining/document.rb'
|
14
14
|
load 'lib/biointerchange/textmining/content.rb'
|
15
|
+
load 'lib/biointerchange/textmining/content_connection.rb'
|
15
16
|
load 'lib/biointerchange/textmining/process.rb'
|
16
17
|
$VERBOSE = v
|
17
18
|
|
@@ -28,23 +29,55 @@ describe BioInterchange::TextMining::PubannosJsonReader do
|
|
28
29
|
model.should be_an_instance_of BioInterchange::TextMining::Document
|
29
30
|
end
|
30
31
|
it 'read json from file' do
|
31
|
-
model = @reader.deserialize(File.new('examples/pubannotation.json'))
|
32
|
+
model = @reader.deserialize(File.new('examples/pubannotation.10096561.json'))
|
32
33
|
|
33
34
|
model.should be_an_instance_of BioInterchange::TextMining::Document
|
34
35
|
end
|
36
|
+
|
37
|
+
it 'read old json from file' do
|
38
|
+
model = @reader.deserialize(File.new('examples/pubannotation.10096561.json.old'))
|
39
|
+
|
40
|
+
model.should be_an_instance_of BioInterchange::TextMining::Document
|
41
|
+
end
|
42
|
+
end
|
43
|
+
|
44
|
+
describe 'old json generated model checks' do
|
45
|
+
before :all do
|
46
|
+
reader = BioInterchange::TextMining::PubannosJsonReader.new("TestOld", "http://test.com", "00-00-0000", BioInterchange::TextMining::Process::UNSPECIFIED, "0.0")
|
47
|
+
|
48
|
+
@model = reader.deserialize('{ "name": "Peter Smith", "name_id": "<peter.smith@example.json>", "date": "2012-08-12", "version": "3", "docurl":"http://example.org/example_json", "text":"Some document text. With two annotations of type protein.\n", "catanns":[{"annset_id":1,"begin":0,"category":"Protein","doc_id":9,"end":10,"id":139},{"annset_id":1,"begin":20,"category":"Protein","doc_id":9,"end":42,"id":138}]}')
|
49
|
+
|
50
|
+
end
|
51
|
+
|
52
|
+
it 'model is of type document' do
|
53
|
+
@model.should be_an_instance_of BioInterchange::TextMining::Document
|
54
|
+
end
|
55
|
+
|
56
|
+
it 'document uri (job id read)' do
|
57
|
+
@model.uri.should eql "http://example.org/example_json"
|
58
|
+
end
|
59
|
+
|
60
|
+
it 'document has content' do
|
61
|
+
@model.contents.size.should eql 3
|
62
|
+
end
|
63
|
+
|
64
|
+
it 'document document' do
|
65
|
+
@model.contents[0].type.should eql BioInterchange::TextMining::Content::DOCUMENT and @model.contents[0].offset.should eql 0 and @model.contents[0].length.should eql 58
|
66
|
+
end
|
67
|
+
|
68
|
+
it 'document phrase' do
|
69
|
+
@model.contents[1].type.should eql BioInterchange::TextMining::Content::PHRASE and @model.contents[1].offset.should eql 0 and @model.contents[1].length.should eql 10 and
|
70
|
+
|
71
|
+
@model.contents[2].type.should eql BioInterchange::TextMining::Content::PHRASE and @model.contents[2].offset.should eql 20 and @model.contents[2].length.should eql 22
|
72
|
+
end
|
35
73
|
end
|
36
74
|
|
37
|
-
describe 'generated model
|
75
|
+
describe 'basic generated model checks' do
|
38
76
|
|
39
77
|
before :all do
|
40
78
|
reader = BioInterchange::TextMining::PubannosJsonReader.new("Test", "http://test.com", "00-00-0000", BioInterchange::TextMining::Process::UNSPECIFIED, "0.0")
|
41
79
|
|
42
|
-
@model = reader.deserialize('{ "name": "Peter Smith", "name_id": "<peter.smith@example.json>", "date": "2012-08
|
43
|
-
|
44
|
-
#puts "Document Model: #{@model.uri}"
|
45
|
-
# @model.contents.each do |c|
|
46
|
-
# puts "\tContent: #{c.type}, #{c.offset}, #{c.length}"
|
47
|
-
#end
|
80
|
+
@model = reader.deserialize('{ "name": "Peter Smith", "name_id": "<peter.smith@example.json>", "date": "2012-12-08", "version": "3", "docurl":"http://example.org/example_json", "text":"Some document text. With two annotations of type protein.\n", "catanns":[{"id":"T1","span":{"begin":0,"end":10},"category":"NP"},{"id":"T2","span":{"begin":20,"end":42},"category":"NP"}]}')
|
48
81
|
end
|
49
82
|
|
50
83
|
it 'model is of type document' do
|
@@ -65,10 +98,49 @@ describe BioInterchange::TextMining::PubannosJsonReader do
|
|
65
98
|
|
66
99
|
it 'document phrase' do
|
67
100
|
@model.contents[1].type.should eql BioInterchange::TextMining::Content::PHRASE and @model.contents[1].offset.should eql 0 and @model.contents[1].length.should eql 10 and
|
68
|
-
|
69
101
|
@model.contents[2].type.should eql BioInterchange::TextMining::Content::PHRASE and @model.contents[2].offset.should eql 20 and @model.contents[2].length.should eql 22
|
70
102
|
end
|
71
|
-
|
103
|
+
end
|
104
|
+
|
105
|
+
describe 'advanced generated model checks' do
|
106
|
+
|
107
|
+
before :all do
|
108
|
+
reader = BioInterchange::TextMining::PubannosJsonReader.new("Test", "http://test.com", "00-00-0000", BioInterchange::TextMining::Process::UNSPECIFIED, "0.0")
|
109
|
+
|
110
|
+
@model = reader.deserialize(File.new('examples/pubannotation.2626671.json'))
|
111
|
+
end
|
112
|
+
|
113
|
+
it 'model is of type document' do
|
114
|
+
@model.should be_an_instance_of BioInterchange::TextMining::Document
|
115
|
+
end
|
116
|
+
|
117
|
+
it 'document uri (job id read)' do
|
118
|
+
@model.uri.should eql "http://www.ncbi.nlm.nih.gov/pubmed/2626671"
|
119
|
+
end
|
120
|
+
|
121
|
+
it 'document has content' do
|
122
|
+
@model.contents.size.should eql 91
|
123
|
+
end
|
124
|
+
|
125
|
+
it 'document document' do
|
126
|
+
#range as exact length seems to depend on encoding used...
|
127
|
+
@model.contents[0].type.should eql BioInterchange::TextMining::Content::DOCUMENT and @model.contents[0].offset.should eql 0 and ( @model.contents[0].length.should > 2350 or @model.contents[0].length.should < 2360 )
|
128
|
+
end
|
129
|
+
|
130
|
+
it 'document content types and interconnections' do
|
131
|
+
doc = 1
|
132
|
+
sub = 39
|
133
|
+
eq = 62
|
134
|
+
th = 64
|
135
|
+
spec = 87
|
136
|
+
neg = 89
|
137
|
+
@model.contents[doc].type.should eql BioInterchange::TextMining::Content::PHRASE and @model.contents[doc].offset.should eql 9 and @model.contents[doc].length.should eql 10 and
|
138
|
+
@model.contents[sub].type.should eql BioInterchange::TextMining::ContentConnection::SUBCLASS and @model.contents[sub].content1.should eql nil and @model.contents[sub].content2.offset.should eql 9 and
|
139
|
+
@model.contents[eq].type.should eql BioInterchange::TextMining::ContentConnection::EQUIVALENCE and @model.contents[eq].content1.offset.should eql 396 and @model.contents[eq].content2.offset.should eql 386 and
|
140
|
+
@model.contents[th].type.should eql BioInterchange::TextMining::ContentConnection::THEME and @model.contents[th].content1.offset.should eql 32 and @model.contents[th].content2.content2.offset.should eql 9 and
|
141
|
+
@model.contents[spec].type.should eql BioInterchange::TextMining::ContentConnection::SPECULATION and @model.contents[spec].content1.content2.offset.should eql 9 and @model.contents[spec].content2.should eql nil and
|
142
|
+
@model.contents[neg].type.should eql BioInterchange::TextMining::ContentConnection::NEGATION and @model.contents[neg].content1.content2.offset.should eql 426 and @model.contents[neg].content2.should eql nil
|
143
|
+
end
|
72
144
|
end
|
73
145
|
|
74
146
|
end
|
@@ -58,6 +58,17 @@ describe BioInterchange::TextMining::RDFWriter do
|
|
58
58
|
ostream.close
|
59
59
|
istream.read.lines.count.should be > 1
|
60
60
|
end
|
61
|
+
|
62
|
+
it 'full advanced json document' do
|
63
|
+
ostream = StringIO.new
|
64
|
+
reader = BioInterchange::TextMining::PubannosJsonReader.new("Test", "http://test.com", "2012-12-09", BioInterchange::TextMining::Process::UNSPECIFIED, "0.0")
|
65
|
+
|
66
|
+
model = reader.deserialize(File.new('examples/pubannotation.2626671.json'))
|
67
|
+
|
68
|
+
BioInterchange::TextMining::RDFWriter.new(ostream).serialize(model)
|
69
|
+
ostream.close_write
|
70
|
+
ostream.string.lines.count.should > 100
|
71
|
+
end
|
61
72
|
end
|
62
73
|
end
|
63
74
|
|