Computational evidence for functionality of noncoding mouse transcripts

AbstractLarge-scale studies of mammalian genome transcription reveal that a large proportion of the genome is transcribed. It remains an open question whether the identified transcripts are functional. Here, we searched for computational evidence to support the functionality of 34,030 noncoding RNA (ncRNA) transcripts reported by the Fantom3 project. We show that compared to control sets, the Fantom ncRNA transcripts set is more conserved with human and rat. We also demonstrate that homologs of the Fantom ncRNA sequences in human and rat have more matches to ESTs. The conserved subgroup of sequences exhibits elevated expression levels in brain tissues. Finally, on average, the Fantom ncRNA sequences have lower minimal free energy of folding than the control sets. Taken together, these observations suggest that, as a group, the Fantom ncRNA set has properties that are different from random sets. Therefore, many of these transcripts may indeed have biological function