Cited research: PLoS Genet. 6, e1000954 (2010)

An individual's genome sequence is not as constant as is commonly thought, according to a large-scale analysis of human DNA and RNA 'editing' by certain proteins.

These proteins can alter DNA and RNA sequences by removing an amine group from one of the 'letters' of the genetic code, which changes the letter and leads to errors in DNA replication or protein synthesis. To inventory such changes, Alexander Wait Zaranek and Erez Levanon, both at Harvard Medical School in Boston, Massachusetts, and their colleagues compared more than 600 million sequences spanning 10 organisms with the corresponding reference genomes.

The authors identified thousands of human DNA sequences and an expanded number of human and mouse RNA sequences that have probably been edited. They also uncovered a common sequencing error that seems to have infiltrated important databases such as that of the HapMap, a resource documenting human genetic variation. H.L.