Data from: Genome-wide discovery and characterization of maize long non-coding RNAs

Background: Long noncoding RNAs (lncRNAs) are transcripts that are 200 bp
or longer, do not encode proteins, and potentially play important roles in
eukaryotic gene regulation. However, the number, characteristics and
expression inheritance pattern of lncRNAs in maize are still largely
unknown. Results: By exploiting available public ESTs, maize whole genome
sequence annotation and RNA-seq datasets from 30 different experiments, we
identified 20,163 putative lncRNA. Of these lncRNAs, more than 90% are
predicted to be the precursors of small RNAs, while 1,704 are considered
to be high-confidence lncRNAs (HC-lncRNAs). HC-lncRNAs have an average
transcript length of 463 bp and HC-lncRNA genes contain fewer exons than
annotated genes (1.2 exons versus 3.5 for the Filtered Gene Set). Based on
the expression pattern of these lncRNAs in 13 distinct tissues and 105
maize recombinant inbred lines (RILs), more than 50% of the HC-lncRNAs are
expressed in a tissue-specific manner, a result that is supported by
epigenetic marks. Intriguingly, the inheritance of lncRNA expression
patterns in 105 RILs reveals apparent transgressive segregation, and maize
lncRNAs are less affected by cis- than by trans- genetic factors.
Conclusions: We integrate all available transcriptomic datasets to
identify a comprehensive set of lncRNAs, provide a unique annotation
resource of the maize genome and a genome-wide characterization of maize
lncRNAs, and explore the genetic control of their expression using
expression quantitative trait locus mapping.