Identifying Somatic Mutations in Poorly Annotated Genomes Using Next Generation Sequencing

The pace at which genome references are being generated for plants and animal species is rapidly increasing with Next Generation Sequencing technologies.  While this is a major step forward for researchers studying species that previously did not have sequenced genomes, it is only the beginning of the process toward defining the biology underlying the genome.  As long as a reference is available, DNA variants can be readily identified on a genome wide scale, often producing lists of 100s of thousands or even millions of variants.  Comparing genomes from the same individual can than generate list of somatic mutations.  Often those occurring in expressed genes are of the most interest; however, if annotation defining where genes exist within a genome is not available or poorly defined, identifying which somatic mutations might affect protein coding may not be possible.  To address this challenge we will describe a method whereby RNA-Seq can be readily used to identify transcriptionally active regions which creates transcript annotation for un-annotated or enhanced annotation for sparsely annotated organisms.  This annotation can then be used in conjunction with whole genome sequencing to annotate variants as to whether they fall within transcriptionally active regions thus facilitating the identification of somatic mutations in expressed regions of the genome for organisms lacking or with poor functional annotation.