Help identify transposable elements (TE's) in manual curation by bringing together several lines of evidence. The input is mainly an alignment of a consensus TE sequence which has been blasted against the original genome, and aligned. These alignments are often fragmented and gappy.
The program of greatest use is probably reputils html. It brings together all of the functionality so far implemented into a single html document.
Building requires Rust.
git clone https://github.com/tolkit/reputils
cd reputils
cargo build --release
# ./target/release/reputils is the executable
# show help
./target/release/reputils --helpEverything is pretty much printed to stdout or to a PNG.
reputils 0.2.0
Max Brown <[email protected]>
reputils - some functions to aid TE identification.
USAGE:
reputils [SUBCOMMAND]
FLAGS:
-h, --help Prints help information
-V, --version Prints version information
SUBCOMMANDS:
con Make a consensus out of a multiple alignment fasta. Optimised for TE's.
div Calculate diversity along sliding windows of an alignment.
dot Make (self) dotplots from a fasta file. Suitable really only for short(ish) sequences.
help Prints this message or the help of the given subcommand(s)
html Render an HTML to gather several lines of identification evidence for a TE.
tir Take a consensus and quickly check for terminal inverted repeats (TIR)
tsd Try to find the Target Site Duplication of a TE. Prints a table.
ttc Trim an alignment to the core TE sequence.
Combines all of the other functions so far into a single HTML document. See the /examples folder for an example. To see it in action, go here.
reputils-html
Render an HTML to gather several lines of identification evidence for a TE.
USAGE:
reputils html --div_window_size <div_window_size> --div_window_step <div_window_step> --dot_nmatch <dot_nmatch> --dot_wsize <dot_wsize> --dot_wstep <dot_wstep> --fasta <fasta> --trim_extend <trim_extend> --trim_iden <trim_iden> --trim_miss <trim_miss> --trim_next_hit <trim_next_hit> --tsd_len <tsd_len> --tsd_min_window <tsd_max_window> --tsd_min_window <tsd_min_window>
FLAGS:
-h, --help Prints help information
-V, --version Prints version information
OPTIONS:
--div_window_size <div_window_size> The size of the window to iterate over. [default: 10]
--div_window_step <div_window_step> The step size of the window to iterate over. If equal to window, then
windows are non-overlapping. [default: 3]
--dot_nmatch <dot_nmatch> Number of matches to tolerate a positive match. [default: 1]
--dot_wsize <dot_wsize> Window size to iterate over sequence. [default: 10]
--dot_wstep <dot_wstep> Window step size for window iterator. [default: 3]
-f, --fasta <fasta> The multiple alignment file in fasta format.
--trim_extend <trim_extend> Extend alingment either end by number of bases specified. [default: 30]
--trim_iden <trim_iden> % identity in a column for the column to be considered a hit. [default:
0.85]
--trim_miss <trim_miss> % missing data tolerated in a column. [default: 0.1]
--trim_next_hit <trim_next_hit> Isolated hits of well conserved columns leads to bad trimming. Play with
this number? [default: 1]
--tsd_len <tsd_len> Number of bases from beginning or end of alignment to query. [default:
30]
--tsd_min_window <tsd_max_window> TSD's are searched for <= to this length. [default: 12]
--tsd_min_window <tsd_min_window> TSD's are searched for >= to this length. [default: 2]
reputils-con
Make a consensus out of a multiple alignment fasta. Optimised for TE's.
USAGE:
reputils con [FLAGS] [OPTIONS] --fasta <fasta>
FLAGS:
-a, --append Append the consensus to the input fasta.
-h, --help Prints help information
-V, --version Prints version information
OPTIONS:
-f, --fasta <fasta> The multiple alignment file in fasta format.
-n, --name <name> Name of the consensus sequence header. [default: CONS]
It's more clear when the alignment is trimmed to include only the putative TE.
reputils-div
Calculate diversity along sliding windows of an alignment.
USAGE:
reputils div [FLAGS] [OPTIONS] --fasta <fasta> --step <step> --window <window>
FLAGS:
-h, --help Prints help information
-p, --plot Plot the diversity across windows of the alignment. Output is a PNG.
-V, --version Prints version information
OPTIONS:
-d, --dir <dir> Directory to put plot in. [default: .]
-f, --fasta <fasta> The consensus sequence file in fasta format.
-n, --name <name> Name of the plot/PNG. [default: div_plot]
-s, --step <step> The step size of the window to iterate over. If equal to window, then windows are non-
overlapping. [default: 25]
-w, --window <window> The size of the window to iterate over. [default: 25]
Quickly check whether your consensus sequence has TIR's.
reputils-tir
Take a consensus and quickly check for terminal inverted repeats (TIR)
USAGE:
reputils tir [FLAGS] --fasta <fasta>
FLAGS:
-h, --help Prints help information
-s, --show Pretty print the alignment.
-V, --version Prints version information
OPTIONS:
-f, --fasta <fasta> The consensus sequence file in fasta format.
This script will take an alignment and trim it to the TE, plus any TSD's (hopefully). It needs a bit of testing, but worked on the Mariners I was looking at. TE's with 5' truncation may not work with this.
reputils-ttc
Trim an alignment to the core TE sequence.
USAGE:
reputils ttc --extend <extend> --fasta <fasta> --identity <identity> --missing <missing> --next_hit <next_hit>
FLAGS:
-h, --help Prints help information
-V, --version Prints version information
OPTIONS:
-e, --extend <extend> Extend the extracted alignment by `e` many bases either side of the alignment.
[default: 15]
-f, --fasta <fasta> The multiple alignment sequence file in fasta format.
-i, --identity <identity> % identity in a column for the column to be considered a hit. [default: 0.8]
-m, --missing <missing> % missing data tolerated in a column. [default: 0.1]
-n, --next_hit <next_hit> Isolated hits of well conserved columns leads to bad trimming. Play with this number?
[default: 1]
Looks at the either end of a trimmed alignment (must be trimmed). I don't know how useful this actually is (it might confuse things more). But here it is:
reputils-tsd
Try to find the Target Site Duplication of a TE. Prints a table.
USAGE:
reputils tsd --fasta <fasta> --length <length> --maximum <maximum> --minimum <minimum>
FLAGS:
-h, --help Prints help information
-V, --version Prints version information
OPTIONS:
-f, --fasta <fasta> The multiple alignment sequence file in fasta format.
-l, --length <length> Number of bases from beginning and end of alignment to query. [default: 20]
-x, --maximum <maximum> TSD's are searched for <= to this length. [default: 12]
-m, --minimum <minimum> TSD's are searched for >= to this length. [default: 2]
Takes a fasta file and self compares each sequence.
reputils-dot
Make (self) dotplots from a fasta file. Suitable really only for short(ish) sequences.
USAGE:
reputils dot --dir <dir> --fasta <fasta> --nmatches <nmatches> --wsize <wsize> --wstep <wstep>
FLAGS:
-h, --help Prints help information
-V, --version Prints version information
OPTIONS:
-d, --dir <dir> Dirname where output plots should go. [default: dot]
-f, --fasta <fasta> The multiple alignment sequence file in fasta format.
-n, --nmatches <nmatches> Number of matches to tolerate a positive match. [default: 1]
-i, --wsize <wsize> Window size to iterate over sequence. [default: 10]
-t, --wstep <wstep> Window step size for window iterator. [default: 4]
Performance will take a dip with large sequences (>10/100kb), especially reputils dot. If you want massive dotplots, there are many other more efficient programs out there!
- simple fasta stats of consensus sequences, sequence length, and length distribution (if multiple fastas)?
- Detect 5' truncation?