P0010 Rapid and Efficient Methods for Preparing rRNA-Depleted and Directional Plant RNA-Seq Libraries

Roy Sooknanan , Epicentre (an Illumina® company), Madison, WI
Cindi A. Hoover , DOE Joint Genome Institute, Walnut Creek, CA
Blake C. Meyers , University Delaware, Newark, DE
John Hitchen , Epicentre (an Illumina® company), Madison
Cristine Kinross , Epicentre (an Illumina® company), Madison, WI
Massively parallel sequencing of cDNA libraries (RNA-Seq) is rapidly gaining momentum for transcript profiling and analysis of novel transcripts, novel isoforms, alternative splice sites, rare transcripts and cSNPs, compared to microarrays. Methods for making RNA-Seq libraries typically comprise preparing rRNA-depleted RNA followed by RNA fragmentation, cDNA synthesis, 5’ and 3’ adaptor-ligation, and multiple clean-up steps. These methods are generally time-consuming, requiring about 2 days with significant hands-on time. Further, after removal of rRNA from total RNA samples using some commercially available kits, the rRNA content can still be very high (>10%-50% of total reads), decreasing mRNA depth and coverage. Here, we present an efficient “single-pass” rRNA removal method (Ribo-Zero™ technology) and an improved, more user-friendly version of the ScriptSeq™ RNA-Seq library preparation method.  Directional cDNA libraries with ~99% strandedness can be prepared in about 2.5 hours, in a single-tube workflow, from either intact or fragmented rRNA-depleted plant total RNA samples, without the need for end-polishing, adaptor-ligation, cDNA fragmentation, or gel-size selection. The combined Ribo-Zero™ and ScriptSeq™ workflow to generate cluster-ready NGS libraries can be completed in about 5 hours. NGS libraries from intact or fragmented total RNA samples contain <2 % of reads that map to rRNA sequences while maintaining representation of coding and noncoding transcripts, independent of polyadenylation. This reduction in rRNA sequence reads improves sequence depth and coverage of mRNA, and increases the percentage of uniquely mapped reads.