Hybrid Assembly Pipeline

This pipeline is for hybrid assembly of microbial genomes using ONT long read and Illumina PE data.

Assumptions

This pipeline assumes you have adapter trimmed and filtered the ONT long reads (min length > 1000bp).
Additionaly, this pipeline assumes you have pre-treated your Illumina reads, but note they will be further kmer corrected using BayesHammer.

Overview

1. Module: SPAdes k-mer Correction

   • SPAdes BayesHammer error correction of Illumina PE data.
   • SPAdes k-mer corrected reads can be found in SPAdes_Cor/corrected/ for alignment.

2. Module: Iterative LoRDEC Long-Read k-mer Correction with Increasing Kmer Size

   • LoRDEC error correction of ONT long read data with kmer corrected Illumina PE reads.
   • LoRDEC corrected long reads remain available for alignment in the LoRDEC subdirectory.

3. Module: Flye de novo Assembly with Long-Reads

   • de novo assembly with k-mer corrected ONT reads to generate assembly graph for Unicycler.
   • Flye iteratively polishes final assembly which remains available in the Flye subdirectory (expect SNVs and miss-assemblies).

4. Module: Final Unicycler Hybrid Assembly

   • Unicycler SPAdes de novo assembly of Illumina reads is closed/corrected with Flye de novo assembly graph.
   • Unicycler built in Pilon polish with chromosomal rotation and iteration corrects consensus assembly.

Dependencies

Pipeline requires the following environments be created.

conda create -n HybridAsBro python=3.6  
source activate HybridAsBro  
conda install spades lordec flye unicycler  
conda deactivate  

Additionally, you will need to increase the memmory usage allowed by Pilon.
To do so, modify the following line in '/home/<user>/miniconda3/env/HybridAsBro/bin/pilon'

default_jvm_mem_opts = ['-Xms512m', '-Xmx1g']

To

default_jvm_mem_opts = ['-Xms1g', '-Xmx12g']

Usage

./HybridAssembly.sh [SAMPLEID].chop.filt.fastq Illumina.R1.fastq.gz Illumina.R2.fastq.gz

Citations

TBD