Home

What is Post-Transcriptional Exon Shuffling (PTES)?

Pre-mRNA transcripts are spliced to remove introns and sequentially join exons in an order consistent with the genomic template. However, recent reports have established the existence of novel transcripts with shuffled exon order different from the genomic context and in the absence of an underlying genomic rearrangement (depicted below). The observation of these novel transcripts has been termed Post-Transcriptional Exon Shuffling (PTES) and results in linear and circular transcripts with little known functional relevance. They are reported to be mostly cytoplasmic, can have expression patterns comparable to canonical transcripts and have been identified in numerous mammalian cell lines, fungi and protists. Despite these reports, our understanding of the structure, origin, distribution and functional relevance remains poor.

What is PTESFinder?

PTESFinder is a computational pipeline for identifying Post-transcriptional Exon Shuffling events from high-throughput RNAseq data. PTESFinder leverages the power of established RNASeq tools and systematically excludes all known classes of false positive structures by applying stringent filters designed to specifically target these false positives. PTESFinder compares alignment qualities of reads mapping to putative PTES structures with qualities of the same reads when mapped to genomic regions and canonically spliced transcripts. This approach increases the confidence in PTES supporting reads. Reads emanating from template-switching events are often characterised by large indels when aligned to the transcriptome. PTESFinder uses additional filters to exclude reads with ambiguous alignments around PTES exon-exon junctions, further increasing confidence in these supporting reads. PTESFinder identifies more PTES structures than other published methods whilst maintaining high specificity.

Memory Requirement:

PTESFinder can analyse RNASeq datasets (with over 300 million reads) within 8G of memory.

Run Times:

PTESFinder can analyze an RNAseq sample (with over 300 million reads) in less than 20 hours, significantly shorter than existing tools.

Overview