Using secondary structure to identify ribosomal numts: cautionary examples from the human genome.

Journal Article (Journal Article)

The identification of inadvertently sequenced mitochondrial pseudogenes (numts) is critical to any study employing mitochondrial DNA sequence data. Failure to discriminate numts correctly can confound phylogenetic reconstruction and studies of molecular evolution. This is especially problematic for ribosomal mtDNA genes. Unlike protein-coding loci, whose pseudogenes tend to accumulate diagnostic frameshift or premature stop mutations, functional ribosomal genes are not constrained to maintain a reading frame and can accumulate insertion-deletion events of varying length, particularly in nonpairing regions. Several authors have advocated using structural features of the transcribed rRNA molecule to differentiate functional mitochondrial rRNA genes from their nuclear paralogs. We explored this approach using the mitochondrial 12S rRNA gene and three known 12S numts from the human genome in the context of anthropoid phylogeny and the inferred secondary structure of primate 12S rRNA. Contrary to expectation, each of the three human numts exhibits striking concordance with secondary structure models, with little, if any, indication of their pseudogene status, and would likely escape detection based on structural criteria alone. Furthermore, we show that the unwitting inclusion of a particularly ancient (18-25 Myr old) and surprisingly cryptic human numt in a phylogenetic analysis would yield a well-supported but dramatically incorrect conclusion regarding anthropoid relationships. Though we endorse the use of secondary structure models for inferring positional homology wholeheartedly, we caution against reliance on structural criteria for the discrimination of rRNA numts, given the potential fallibility of this approach.

Full Text

Duke Authors

Cited Authors

  • Olson, LE; Yoder, AD

Published Date

  • January 2002

Published In

Volume / Issue

  • 19 / 1

Start / End Page

  • 93 - 100

PubMed ID

  • 11752194

Electronic International Standard Serial Number (EISSN)

  • 1537-1719

International Standard Serial Number (ISSN)

  • 0737-4038

Digital Object Identifier (DOI)

  • 10.1093/oxfordjournals.molbev.a003987


  • eng