Very fast identification of RNA motifs in genomic DNA. Application to tRNA search in the yeast genome

J Mol Biol. 1996 Nov 22;264(1):46-55. doi: 10.1006/jmbi.1996.0622.


A common strategy characterises the various methods independently defined to identify almost unambiguously different types of RNA molecules in DNA fragments. So far, the good quality of detection of RNA motif has been the prior motivation and effectively delayed the optimisation of programs. As an illustration of possible improvements, a modified version of tRNAscan is described. The previous algorithm was altered to run 500 times faster and to lower both rates of false positives and false negatives. The newly sequenced genome of Saccharomyces cerevisiae is scanned both ways in less than three minutes and results match annotations found in databanks with three exceptions, two of which being arguably not real tRNAs.

Publication types

  • Comparative Study

MeSH terms

  • Algorithms
  • Animals
  • Base Sequence
  • Consensus Sequence
  • DNA, Fungal / genetics*
  • Databases, Factual
  • Evaluation Studies as Topic
  • Genome, Fungal*
  • Molecular Sequence Data
  • RNA, Fungal / genetics*
  • RNA, Transfer / genetics*
  • Saccharomyces cerevisiae / genetics*
  • Sensitivity and Specificity
  • Sequence Alignment / methods
  • Sequence Alignment / statistics & numerical data
  • Software


  • DNA, Fungal
  • RNA, Fungal
  • RNA, Transfer