INC-Seq: Accurate single molecule reads using nanopore sequencing

Description:

This repository contains the code for analyzing INC-Seq data (http://biorxiv.org/content/early/2016/01/27/038042). The full datasets have been deposited into ENA (http://www.ebi.ac.uk/ena/data/view/PRJEB12294).

Note:

The PBDAGCON software attached with this pipeline was compiled on Ubuntu 16.04. If there is any issue with the consensus building, please recompile PBDAGCON. On Debian systems, this can be done by running:

rm -i utils/pbdagcon
sudo apt install pbdagcon
ln -s `which pbdagcon` utils/

Requirements:

• Python 2.7
• Biopython 1.65
• BLAST 2.2.28+

Usage:

usage: inc-seq.py [-h] -i INFASTA [-o OUTFILE] [-a ALIGNER] [-m MINRL]
                  [--anchor_seg_step ANCHOR_SEG_STEP]
                  [--anchor_length ANCHOR_LEN] [--anchor_cov ANCHOR_COV]
                  [--anchor_seq ANCHOR_SEQ] [--iterative] [--seg_cov SEG_COV]
                  [--copy_num_thre COPY_NUM_THRE]
                  [--length_difference_threshold LEN_DIFF_THRE]

The INC-Seq pipeline

optional arguments:
  -h, --help            show this help message and exit
  -i INFASTA, --input INFASTA
                        Input file in fasta format
  -o OUTFILE, --outfile OUTFILE
                        Output file
  -a ALIGNER, --aligner ALIGNER
                        The aligner used (blastn, graphmap, poa) [Default:
                        blastn]
  -m MINRL, --minReadLength MINRL
                        The reads shorter than this will be discarded
                        [Default:2000]
  --anchor_seg_step ANCHOR_SEG_STEP
                        Step of sliding window used as anchors [Default: 500]
                        (eg. -s 500 : start at 0, 500, 1000, ...)
  --anchor_length ANCHOR_LEN
                        The length of the anchor, should be smaller than the
                        unit length [Default: 500]
  --anchor_cov ANCHOR_COV
                        Anchor coverage required [Default: 0.8]
  --anchor_seq ANCHOR_SEQ
                        A single file containing the sequences used as the
                        anchor [Default: Use subsequences as anchors]
  --iterative           Iteratively run pbdagcon on consensus [Default: False]
  --seg_cov SEG_COV     Segment coverage required [Default: 0.8]
  --copy_num_thre COPY_NUM_THRE
                        Minimal copy number required [Default: 6]
  --length_difference_threshold LEN_DIFF_THRE
                        Segment length deviation from the median to be
                        considered as concordant [Default: 0.05]

Examples:

• Basic usage

./inc-seq.py -i data/inc_seq_test_read.fa -o consensus.fa

• Use graphmap as segment aligner

./inc-seq.py -i data/inc_seq_test_read.fa -o consensus.fa -a graphmap

• Use bpipe pipeline for pseudo-parallel computing
• Split the reads into multiple files (300 reads per file) and run INC-Seq (4 instances) in parallel.

bpipe run -p READ_NUM=300 -n 4 pipeline.bpipe a_lot_of_incseq_reads.fa