Consensus bacterial genome assembly pipeline that utilizes the Flye assembler (https://github.com/fenderglass/Flye) with downstream short-read and long-read correction tools to provide contiguous, circularized replicons with low error rates.
- Flye (≥ 2.7)
- BEDtools (≥ 2.29.2)
- VCFtools (≥ 0.1.15)
- berokka (≥ 0.2)
- perl (≥ 5.28.0)
- ncbi-blast+ (≥ 2.10.0)
- Python (≥ 3.6.0)
- Medaka (≥ 0.11.5)
- MUMmer (≥ 3.23)
- Prodigal (≥ 2.6.3)
- Minimap2 (≥ 2.17-r974)
- Racon (≥ 1.4.5)
- bwa (≥ 0.7.17)
- SAMtools (≥ 1.10)
- Parallel (≥ 201807022)
- BCFtools (≥ 1.10.2)
- htslib (≥ 1.10.2)
Versions above have been tested. Older or newer versions may work, but could create conflicts. Suggestion is to create a virtual environment with these dependencies.
Required input are: (1) out-directory (outdir) (2) ONT long-reads (either gzip or non-compressed files will work) (3) Interleaved paired-end short-reads (either gzip or non-compressed files will work) (4) genome_size estimated genome size based on previous knowledge of species (+/- 1 Mb genome size estimate is fine for flye k-mer selection)
Note that paired-end reads MUST be interleaved for racon to work. You can interleave short-read data using the bbmap tool reformat.sh. For racon to work properly, make sure there are underscores in lieu of white-space in headers of short-read fastq files by using the underscore=t option in reformat.sh. Example command-line for reformat.sh is as follows:
$reformat.sh in1=PE_read1.fastq.gz in2=PE_read2.fastq.gz out=PEIL.fastq.gz underscore=t
Usage with Flye assembler:
$python3 flye_pipeline.py -t 2 -s sample_name -o outdir -pe interleaved_pe_reads.fastq.gz -l long_reads.fastq.gz --genome_size 5.3m -d dnaA_file.fasta
Usage for error correction of previously assembled genome:
$python3 flye_pipeline.py -t 2 -s sample_name -o outdir -x -c assembly.fasta -pe interleaved_pe_reads.fastq.gz -l long_reads.fastq.gz -d dnaA_file.fasta
Final assembly can be found in outdir_name/shortRead_polish_results/ and will be named SAMPLENAME_final.fasta
(1) Note that any database of genes of interest to orient contigs can be used for the purposes of orienting any linear or circular DNA structure when using the -d paramter in pipeline. For example, if assembling K. pneumoniae genomes, one can use the manually curated, database of commonly observed F-type replication initiation protein genes found in plasmids as well as the dnaA gene for K. pneumoniae in the db directory with the file name dnaA_and_plasmid_startSites.fasta.
If one wants to use --meta and --plasmid options for their flye assemblies, please use the -mp argument in the pipeline.