Fig. 3

Strategy for bioinformatic analyses of whole genome sequencing (WGS) data to identify Tnt1 insertion sites. a Tobacco (Nicotiana tabacum) type I retrotransposon (Tnt1) genomic structure. Tnt1 transposon sequences contain 610 bp long terminal repeats (LTRs) on both left (LE) and right ends (RE), and encode a capsid protein (GAG), protease (PR), an integrase (INT), reverse transcriptase (RT) and RNAseH (RH). Paired-end (PE) sequencing reads obtained from Tnt1 insertion mutants are classified into 5 different types: Tnt1-Tnt1 (type 1), genomic-genomic (type 2), hybrid-Tnt1 or Tnt1-hybrid (type 3), genomic-hybrid or hybrid-genomic (type 4) and genomic-Tnt1or Tnt1-genomic (type 5) based on the sequence composition. Read types 3, 4, and 5 were used in our analyses. b Workflow for the identification of Tnt1 insertion loci using the WGS data. To identify the Tnt1 insertion loci, a local BLAST program [29] was used. Clean PE reads were aligned to 90 bp LE and RE Tnt1 sequences to identify type 3 reads. These are putative “hybrids” containing Tnt1-genomic sequence junctions. Type 4 and 5 reads were used to develop contigs/nodes using the VELVET program [47] to assemble genomic fragments. Both hybrid reads and contigs/nodes were aligned to A17 reference genome by BLAST to obtain genomic co-ordinates of putative Tnt1 insertion loci