Skip to main content

Table 7 Two sentences extracted from the abstract in Table 1 are assigned a ranked set of N Gene Ontology descriptors (N = 2). Each {sentence;category} association pair is provided with a categorization status value (CSV), which directly expresses a similarity between the sentence and the Gene Ontology. The final density is computed by simply summing up the top-N CSV (N = 2).

From: Gene Ontology density estimation and discourse analysis for automatic GeneRiF extraction

Sentences Predicted GO categories CSV Density
Class 2 transcripts predominated in most other tissues rna primary transcript binding 1028 2056
  35s primary transcript processing 1028  
regulation of ABCA1 mRNA levels exploits the use of alternative transcription start sites transcription 9397 12819
  regulation 3422