Skip to content

Assembly

Description: Launches assemblers (user selected from SPAdes, MEGAHIT and Penguin, default is all), compute basic assembly statistics for each, merges the different assemblers contigs via linclust (i.e. reduces duplicates and contigs from one assembler wholly contained within a contigs from another tool), and finally filters the merged contigs based on user-defined parameters.

flowchart TD
    subgraph subGraph0["<b>Input reads</b>"]
        B["<i>supported input types (gzip allowed)</i>:<br> 
            • interleaved fastq files<br> 
            • merged fastq files<br> 
            • directory of fastq files"]
    end

    subgraph subGraph1["<b>Assembly</b>"]
        A1["SPAdes"]
        A2["MEGAHIT"]
        A3["Penguin"]
    end
T1["<b> Initial Assemblies </b>"]
    subgraph subGraph2["<b>Post-Processing</b>"]
        P1["Deduplication<br> <i>(seqkit rmdup)</i>
        <i>(combined contigs)</i>"]
        P2["Read Mapping<br> <i>(BBWrap + Bowtie1)</i>"]
        P1 --> P2

    end

    subgraph subGraph3["<b>Outputs</b>"]
        O2["Coverage Statistics<br> <i>(covstats.txt, covhist.txt)</i>"]
        O3["Mapped Reads<br> <i>(*.sam.gz files)</i>"]
        O4["Unassembled Reads<br> <i>(bbw_unassembled.fq.gz)</i>"]
    end

    B --> subGraph1
    subGraph1 --> T1
    T1 --> subGraph2
    subGraph0 --> P2
    P2 --> O3
    P2 --> O2
    P2 --> O4

Options

Common

  • -id, --input-dir: Input fastq files or directory (required)
  • -o, --output: Output directory (default: "RP_assembly_output")
  • -t, --threads: Number of threads (default: 1)
  • -M, --memory: RAM limit (default: "6gb")
  • -g, --log-file: Path to log file (default: current_directory/assemble_logfile.txt)
  • --keep-tmp: Keep temporary files (flag)

Assembly

  • -A, --assembler: Assemblers to use (default: "spades,megahit,penguin")
  • --override-parameters: JSON-like string to override defaults
  • --skip-steps: Steps to skip (comma-separated)
  • --overwrite: Overwrite existing output (flag)

Default Parameters

Note: kmer lengths below are the max and min values Rolypoly will try, but it may reduce the range based on the data,to avoid errors (e.g. kmers longer the provided reads, too many kmers, or kmers of even number (if user overrides the default)).

SPAdes

  • k-mer sizes: 21,33,45,57,63,69,71,83,95,103,107,111,119
  • mode: meta

MEGAHIT

  • k-min: 21
  • k-max: 147
  • k-step: 8
  • min-contig-len: 30

Penguin

  • min-contig-len: 150
  • num-iterations: aa:1,nucl:12

BBWrap

  • maxindel: 200
  • minid: 90
  • untrim: true
  • ambig: best

Skipable steps:

  • -ss, --skip-steps: Steps to skip (comma-separated).
    options:
  • seqkit: Deduplication (seqkit rmdup)
  • bbwrap: for read mapping and assembly stats
  • bowtie: for read mapping (sams).

Downstream applications:

You can use rolypoly filter-contigs to remove contigs mapped to specific reference genome. This is useful for host DNA decontamination.

Citations

Assemblers

  • SPAdes: Genome assembler

  • Citation: https://doi.org/10.1089/cmb.2012.0021

  • MEGAHIT: Ultra-fast metagenome assembler

  • Citation: https://doi.org/10.1093/bioinformatics/btv033

  • Penguin: Virus-aware assembler

  • Citation: https://doi.org/10.1101/2024.03.29.587318

Support Tools

  • BBMap: Read mapping and processing

  • Citation: https://sourceforge.net/projects/bbmap/files/BBMap_39.08.tar.gz

  • SeqKit: Sequence manipulation

  • Citation: https://doi.org/10.1002/imt2.191

  • Bowtie: Read mapping

  • Citation: https://doi.org/10.1186/gb-2009-10-3-r25