OSIRISv1.2: A named entity recognition system for sequence variants of genes in biomedical literature

BMC Bioinformatics

Table 2 Detail of ambiguous annotations by OSIRISv1.2

The table show the cases where an ambiguous assignment of a dbSNP identifier to a variation term was obtained. The first column depicts the PMID of the abstracts where the term (second column) was found. The third column shows the dbSNP identifiers assigned to the term by OSIRISv1.2. Highlighted in bold are the correct assignment of dbSNP identifier to that term. In some instances the information from the text does not allow to discern between two SNPs occurring at the same position in the protein sequence but differing only in one of the alternative residues. For example, for the "Asp86" term, both SNPs highlighted in bold represent changes at position 86 of the protein in which an Asp residue is involved. One is a synonymous change while the other is non-synonymous. However, the information extracted from the text is not sufficient to distinguish among these two possibilities (only one allele is described: the Asp residue), and therefore both identifiers are assigned to the term.

ISSN: 1471-2105