Module: Annotations
- Defined in:
- lib/MARQ/annotations.rb
Defined Under Namespace
Constant Summary collapse
- RANK_SIZE_BINS =
%w(1 2 3 4 5 7 10 15 20 30 40 50 65 80 100 125 150 175 200 250 300 350 400 450 500 600 700 800 900 1000 1500 2000 2500 3000)
- @@terms_cache =
{}
Class Method Summary collapse
- .annotations(scores, type, pvalue = 0.05, algorithm = :rank) ⇒ Object
- .compare(a, b) ⇒ Object
- .dataset_annotations(dataset, type, experiment) ⇒ Object
- .enrichment_hypergeometric(annotations, relevant, options) ⇒ Object
- .enrichment_rank(annotations, ranks, options = {}) ⇒ Object
- .exp2gds(experiment) ⇒ Object
Class Method Details
.annotations(scores, type, pvalue = 0.05, algorithm = :rank) ⇒ Object
302 303 304 305 306 307 308 309 310 311 312 313 314 315 316 317 318 319 320 321 322 323 324 325 326 327 328 329 330 331 332 333 334 335 336 337 338 339 340 341 342 343 344 345 346 347 348 349 350 351 352 353 |
# File 'lib/MARQ/annotations.rb', line 302 def self.annotations(scores, type, pvalue = 0.05, algorithm = :rank) annot = {} relevant = [] case when type =~ /^(.*)_direct$/ side = :direct type = $1 when type =~ /^(.*)_inverse$/ side = :inverse type = $1 end scores.each{|experiment, info| dataset = experiment.match(/^(.*?): /)[1] name = $'.strip case when side.nil? experiment_type = type when side == :direct && info[:score] >= 0 || side == :inverse && info[:score] < 0 experiment_type = type + '_up' else experiment_type = type + '_down' end annot[experiment] = dataset_annotations(dataset, experiment_type, name) relevant << experiment if info[:pvalue] <= pvalue } = {} if type == "Words" = {:low => 0, :hi => 0.05, :limit => 100000} else = {:low => 0, :hi => 0.5, :limit => 100000} end if algorithm == :rank ranks = scores.sort {|a,b| compare(a[1],b[1]) }.collect {|p| p[0]} terms = enrichment_rank(annot, ranks, ) else terms = enrichment_hypergeometric(annot, relevant, ) end merged_annotations = {} annot.each{|key, info| merged_annotations[key] = info[:dataset] + info[:signature] } [merged_annotations, terms] end |
.compare(a, b) ⇒ Object
104 105 106 107 108 109 110 111 112 113 |
# File 'lib/MARQ/annotations.rb', line 104 def self.compare(a,b) case when a[:pvalue] < b[:pvalue] -1 when a[:pvalue] > b[:pvalue] 1 when a[:pvalue] == b[:pvalue] b[:score].abs <=> a[:score].abs end end |
.dataset_annotations(dataset, type, experiment) ⇒ Object
288 289 290 291 292 293 294 295 296 297 298 299 300 |
# File 'lib/MARQ/annotations.rb', line 288 def self.dataset_annotations(dataset, type, experiment) annotation_dir = File.join(MARQ.datadir, (MARQ::Dataset.is_GEO?(dataset) ? 'GEO' : 'CustomDS'), 'annotations') term_file = File.join(annotation_dir, type, MARQ::Name.clean(dataset)) if File.exist? term_file @@terms_cache[term_file] ||= YAML::load(File.open(term_file)) terms = @@terms_cache[term_file] {:dataset => (terms[:dataset] || []), :signature => (terms[experiment] || [])} else {:dataset => [], :signature => []} end end |
.enrichment_hypergeometric(annotations, relevant, options) ⇒ Object
220 221 222 223 224 225 226 227 228 229 230 231 232 233 234 235 236 237 238 239 240 241 242 243 244 245 246 247 248 249 250 251 252 253 254 255 256 257 258 259 260 261 262 263 264 265 266 267 268 269 270 271 272 273 274 275 276 277 278 279 280 281 282 283 284 |
# File 'lib/MARQ/annotations.rb', line 220 def self.enrichment_hypergeometric(annotations, relevant, ) = { :dict_options => {:low => 0, :hi => 0.5, :limit => 100000} }.merge()[:dict_options] positions = {} found_datasets = [] dict = Dictionary::TF_IDF.new ranks.each_with_index{|experiment, rank| info = annotations[experiment] dataset_terms = info[:dataset] signature_terms = info[:signature] dataset = exp2gds experiment terms = signature_terms terms += dataset_terms term_count = {} terms.each{|term| term_count[term] ||= 0 term_count[term] += 1 } dict.add(term_count) } best = dict.best().keys terms = {} found_datasets = [] annotations.each{|experiment, info| dataset_terms = info[:dataset] signature_terms = info[:signature] dataset = exp2gds experiment signature_terms.each{|term| next if ! best.include? term terms[term] ||= {:relevant => 0, :total => 0} terms[term][:total] += 1 terms[term][:relevant] += 1 if relevant.include? experiment } next if found_datasets.include? dataset found_datasets << dataset dataset_terms.each{|term| next if ! best.include? term terms[term] ||= {:relevant => 0, :total => 0} terms[term][:total] += 1 terms[term][:relevant] += 1 if relevant.include? experiment } } total = annotations.keys.length list = relevant.length terms.each{|term, info| info[:pvalue] = Annotations.hypergeometric(total,info[:total],list, info[:relevant]) } terms end |
.enrichment_rank(annotations, ranks, options = {}) ⇒ Object
117 118 119 120 121 122 123 124 125 126 127 128 129 130 131 132 133 134 135 136 137 138 139 140 141 142 143 144 145 146 147 148 149 150 151 152 153 154 155 156 157 158 159 160 161 162 163 164 165 166 167 168 169 170 171 172 173 174 175 176 177 178 179 180 181 182 183 184 185 186 187 188 189 190 191 192 193 194 195 196 197 198 199 200 201 202 203 204 205 206 207 208 209 210 211 212 213 214 215 216 217 218 |
# File 'lib/MARQ/annotations.rb', line 117 def self.enrichment_rank(annotations, ranks, = {}) = { :dict_options => {:low => 0, :hi => 0.5, :limit => 100000} }.merge()[:dict_options] positions = {} found_datasets = [] dict = Dictionary::TF_IDF.new ranks.each_with_index{|experiment, rank| info = annotations[experiment] dataset_terms = info[:dataset] signature_terms = info[:signature] dataset = exp2gds experiment terms = signature_terms terms += dataset_terms term_count = {} terms.each{|term| term_count[term] ||= 0 term_count[term] += 1 } dict.add(term_count) } best = dict.best().keys found_datasets = [] ranks.each_with_index{|experiment, rank| info = annotations[experiment] dataset_terms = info[:dataset] signature_terms = info[:signature] dataset = exp2gds experiment terms = signature_terms if ! found_datasets.include? dataset terms += dataset_terms found_datasets << dataset end terms.uniq.each{|term| next if not best.include? term positions[term] ||= [] positions[term] << rank } } sizes = {} RANK_SIZE_BINS.each{|size| sizes[size.to_i] = []} # For each term compute the rank score. Also, place it in the closest size # bin for the permutations. scores = [] best.each_with_index{|term, pos| if positions[term] list = positions[term] # place it on the size bin found = false sizes.keys.sort.each_with_index{|size,i| next if found if list.length < size found = true sizes[sizes.keys.sort[i-1]] << pos end } sizes[sizes.keys.sort.last] << pos if !found scores << Score.score(list, ranks.length, 0) else # it has no score scores << nil end } info = {} sizes.each do |size, pos_list| next if size == 1 next if pos_list.empty? size_info = {} pos_list.each do |pos| score = scores[pos] term = best[pos] hits = positions[term].nil? ? 0 : positions[term].length size_info[term] = {:score => score, :hits => hits} end null_scores = Score.null_scores(size, 0) size_info = Score.add_pvalues(size_info, null_scores) info.merge! size_info end info end |
.exp2gds(experiment) ⇒ Object
99 100 101 102 |
# File 'lib/MARQ/annotations.rb', line 99 def self.exp2gds(experiment) experiment =~ /(.*?):/ $1 end |