blue-moon22 / palidis

PaliDIS: A tool for fast discovery of novel insertion sequences

PaliDIS is retired, please refer to PaliDIS4

palidis - Palindromic Detection of Insertion Sequences

Contents

• Introduction
• Description
• Installation
• Usage
• Output

Introduction

Palidis is a tool that discovers novel insertion sequences.

The tool is based upon identifying inverted terminal repeats (ITRs) (figure below) using paired-end, short-read metagenomic data/mixed microbial genomes.

For each sample, palidis produces two output files: 1. FASTA file of insertion sequences and 2. Information for each insertions sequence

Description

Steps:

1. Pre-process FASTQ.GZ reads [convertToFasta]
2. Efficient maximal exact matching to get repeat sequences using pal-MEM [palmem]
3. Map reads against assemblies using Bowtie2 [filterContigs buildDB mapreads]
4. Get candidate ITRs by distance filters [getCandidateITRs]
5. Cluster candidate ITRs using CD-HIT-EST [clusterReads]
6. Get putative ITRs by cluster concordance and output Insertion Sequences [getITRs]
7. Get insertion sequences [runProdigal, installInterproscan, runInterproscan, getISInfo]

Installation

• Install Nextflow
• Install Docker if using own machine or install Singularity/load a singularity module if using a shared HPC
• Clone this repo:

git clone --recursive -j8 https://github.com/blue-moon22/palidis.git
cd palidis

Note: You may be warned to first call git config --global --add safe.directory.
If you have already cloned this repo with git clone https://github.com/blue-moon22/palidis.git, you also need to get the submodules git submodule update --init --recursive

Usage

nextflow palidis.nf --manifest <manifest_file> --batch_name <batch_name> -c configs/conf/<name_of_config>.config

Mandatory arguments

<batch_name>

<batch_name> must be the directory that the output is stored in.

<manifest_file>

A tab-delimited manifest must be specified for --manifest containing the absolute paths with headers lane_id, read1, read2, sample_id and contigs_path, e.g. this manifest contains three samples (the first having two lanes and the other two having one lane):

lane_id	read1	read2	sample_id	contigs_path
lane1	/path/to/file/lane1_1.fq.gz	/path/to/file/lane1_2.fq.gz	my_sample1	/path/to/file/contigs.fasta
lane2	/path/to/file/lane2_1.fq.gz	/path/to/file/lane2_2.fq.gz	my_sample1	/path/to/file/my_sample1_contigs.fasta
lane3	/path/to/file/lane3_1.fq.gz	/path/to/file/lane3_2.fq.gz	my_sample2	/path/to/file/my_sample2_contigs.fasta
lane4	/path/to/file/lane4_1.fq.gz	/path/to/file/lane4_2.fq.gz	my_sample3	/path/to/file/my_sample3_contigs.fasta

<name_of_config>

This represents the institution or HPC name. You can find your institutional HPC's config in configs/conf (which is linked to the configs directory in nf-core). For example, running on Sanger's HPC: -c configs/conf/sanger.config

Optional arguments

  --min_itr_length    Minimum length of ITR. (Default: 25)
  --max_itr_length    Maximum length of ITR. (Default: 50)
  --kmer_length       k-mer length for maximal exact matching. (Default: 15)
  --min_is_len        Minimum length of insertion sequence. (Default: 500)
  --max_is_len        Maximum length of insertion sequence. (Default: 3000)
  --cd_hit_G          -G option for CD-HIT-EST. (Default: 0)
  --cd_hit_aL         -aL option for CD-HIT-EST. (Default: 0.0)
  --cd_hit_aS         -aS option for CD-HIT-EST. (Default: 0.9)
  --cd_hit_c          -c option for CD-HIT-EST. (Default: 0.9)
  -resume             Resume the pipeline

Testing

If you would like to test whether this pipeline produces the expected output on your system, run this command. If successful, it should print Test passed. All outputs expected..

./tests/regression_tests.sh

Output

There are two output files stored in a directory specified with --batch_name:

1. FASTA file of insertion sequences

2. Information for each insertions sequence

IS_name	sample_id	contig	itr1_start_position	itr1_end_position	itr2_start_position	itr2_end_position	description
IS_length_655-IPR002686_154_418-IPR002686_148_565-IPR036515_124_667-IPR036515_124_580-PTHR36966_133_625	SRS013170	NODE_18_length_76504_cov_9.77495	74408	74436	75032	75062	IPR002686:Transposase IS200-like;IPR036515:Transposase IS200-like superfamily;PTHR36966:REP-ASSOCIATED TYROSINE TRANSPOSASE
IS_length_1455-IPR013762_1393_1918	SRS013170	NODE_31_length_64375_cov_7.58579	10034	10063	11459	11488	IPR013762:Integrase-like, catalytic domain superfamily

Interpretation

Header	Description
IS_name	Name assigned by PaliDIS which contains the length, interpro or PANTHER accessions of transposases and their positions, e.g. IS_length_655-IPR002686_154_418-IPR002686_148_565-IPR036515_124_667-IPR036515_124_580-PTHR36966_133_625 represents an IS of nucleotide length 655 with transposases detected including Interpro accession IPR002686 in positions 154-418 and 148-565, Interpro accession IPR036515 in position 124-667 and PANTHER accession PTHR36966 in position 133-625)
sample_id	Sample ID that was given in manifest
contig	Name of the contig that was given by the header in the contig file provided by the manifest
itr1_start_position	The position in the contig of the first nucleotide of the left-hand Inverted Terminal Repeat (ITR) sequence (also the start of the IS)
itr1_end_position	The position in the contig of the last nucleotide of the left-hand ITR sequence
itr2_start_position	The position in the contig of the first nucleotide of the right-hand ITR sequence
itr2_end_position	The position of the last nucleotide of the right-hand ITR sequence (also the end of the IS)
description	The description of each accession recorded in IS_name, e.g. IPR002686:Transposase IS200-like;IPR036515:Transposase IS200-like superfamily;PTHR36966:REP-ASSOCIATED TYROSINE TRANSPOSASE

palidis's Issues

Issue with prodigal not picking up same genes for reverse sequences

Related to issue hyattpd/Prodigal#90

Option to switch to Pyrodigal or a Prodigal patch which I heard exists

input genomes

Hi, thanks for providing palidis.
How would I use this with a list of input genomes already assembled from metagenomes?
Thanks!