Culture-Independent Discovery of Viroids by Deep Sequencing and Computational Algorithms.

Viroids are single-stranded circular RNA molecules that cause diseases in plants and do not encode any protein. Classical approaches for the identification of new viroids are challenging for many plant pathology laboratories as viroid cDNA synthesis and sequencing require purification and enrichment of the naked viroid RNA by two-dimensional gel electrophoresis. Conventional metagenomic approaches are not effective for viroid discovery because the total number of known viroids is small, and distinct viroids share limited nucleotide sequence similarity. In this chapter, we describe a homology-independent approach for the identification of both known and new viroids in disease samples. It is known that viroid infection of plants triggers production of overlapping viroid-derived small interfering RNAs (siRNAs) targeting the entire genome with high densities and that replication of viroids occurs via a rolling-circle mechanism to yield head-to-tail multiple-repeat replicative intermediates. Our approach involves deep sequencing of either long or small RNAs in a disease sample followed by viroid identification with a unique computational algorithm, progressive filtering of overlapping small RNAs (PFOR). Among the sequenced total small RNAs, PFOR retains viroid-derived siRNAs for viroid genome assembly by progressively eliminating nonoverlapping small RNAs and those that overlap but cannot be assembled into a direct repeat RNA, a unique feature of viroid RNA replication. In contrast, long RNAs sequenced after depletion of ribosomal RNAs are cut into 21-nucleotide virtual overlapping small RNAs with the algorithm SLS (splitting longer read into shorter fragments) before PFOR. We show that new viroids or viroids from the two known families are readily identified and their full-length sequences recovered by PFOR from long or small RNAs sequenced directly from infected plants. We propose that our approach can be used for viroid discovery in both plants and potentially animals since PFOR identifies viroids by searching for circular RNAs or a unique replication intermediate of the viroid genome in a sequence homology-independent manner.