STELR (pronounced Stellar) is a re-implementation of the fast non-reference transposable element (TE) detector from long read sequencing data (PacBio or Oxford Nanopore) called TELR that leverages the Snakemake workflow manager. STELR uses long reads mapped to a reference genome to identify insertions using Sniffles, then filters insertions by matching insertion supporting reads with user supplied TE consensus sequences. For each TE insertion candidate locus, STELR performs a local assembly of all reads supporting TE insertion, annotates the TE sequence in assembled contigs, then maps the flanks back to the reference genome. Finally, STELR generates the coordinates of the non-reference TE insertions, the estimated allele frequency and the assembled TE sequences.
The STELR pipeline consists of four main stages: (1) general SV detection and filter for TE insertion candidate, (2) local reassembly and polishing of the TE insertion, (3) identification of TE insertion coordinates, and (4) estimation of intra-sample TE insertion allele frequency.
-
In stage 1, long reads are aligned to the reference genome using NGMLR (https://github.com/philres/ngmlr). The alignment output in BAM format is provided as input for Sniffles (https://github.com/fritzsedlazeck/Sniffles) to detect structural variations (SVs). STELR then filter for TE insertion candidates from SVs reported by Sniffles using following criteria: 1) The type of SV is insertion. 2) Insertion sequence is available. 3) The insertion sequences include hits from user provided TE consensus library using RepeatMasker (http://www.repeatmasker.org}).
-
In stage 2, reads that support the TE insertion candidate locus based on Sniffles output are used as input for wtdbg2 (https://github.com/ruanjue/wtdbg2) or flye (https://github.com/fenderglass/Flye) to assemble local contig that covers the TE insertion for each TE insertion candidate locus. The local assemblies are then polished using wtdbg2 or flye. Note that 1) each assembler can be matched with each polisher, and 2) minimap2 is used to re-align reads to local contig for polishing.
-
In stage 3, TE consensus library is aligned to the assembled TE insertion contigs using minimap2 and used to define TE-flank boundaries. TE region in each contig is annotated with family info using RepeatMasker. Sequences flanking the TE insertion are then re-aligned to the reference genome using minimap2 to determine the precise TE insertion coordinates and target site duplication (TSD).
-
In stage 4, raw reads aligned to the reference genome are extracted within a 1kb interval on either side of the insertion breakpoints initially defined by Sniffles. The reads are then aligned to the assembled polished contig to identify reads that support the TE insertion and reference alleles, respectively, in following steps: 1) Reads are aligned to the forward strand of the contig, 5' flanking sequence depth (
5p_flank_cov) and 5' TE depth (5p_te_cov) are calculated. 2) Reads are aligned to the reverse complement strand of the contig, 5' flanking sequence depth (3p_flank_cov) and 5' TE depth (3p_te_cov) are calculated. 3) The TE allele frequency is estimated as(5p_te_cov/5p_flank_cov + 3p_te_cov/3p_flank_cov)/2.
The current version of STELR shows good performance on real Drosophila melanogaster data sets, including datasets with heterozygous TE insertions.
STELR is written in python3 and is designed to run on linux operating system.
The following sections will provide you installation instructions, usage guide, and descriptions of output files.
Please use the Github Issue page if you have questions.
To cite STELR in publications, please use:
Shunhua Han, Guilherme B Dias, Preston J Basting, Raghuvir Viswanatha, Norbert Perrimon, Casey M Bergman (2022) Local assembly of long reads enables phylogenomics of transposable elements in a polyploid cell line. Nucleic Acids Research 50(21):e124 https://doi.org/10.1093/nar/gkac794.