FASTP

trim and QC fastq reads with fastp

Example

This wrapper can be used in the following way:

rule fastp_se:
    input:
        sample=["reads/se/{sample}.fastq"]
    output:
        trimmed="trimmed/se/{sample}.fastq",
        failed="trimmed/se/{sample}.failed.fastq",
        html="report/se/{sample}.html",
        json="report/se/{sample}.json"
    log:
        "logs/fastp/se/{sample}.log"
    params:
        adapters="--adapter_sequence ACGGCTAGCTA",
        extra=""
    threads: 1
    wrapper:
        "v1.25.0-52-g79342b73/bio/fastp"


rule fastp_pe:
    input:
        sample=["reads/pe/{sample}.1.fastq", "reads/pe/{sample}.2.fastq"]
    output:
        trimmed=["trimmed/pe/{sample}.1.fastq", "trimmed/pe/{sample}.2.fastq"],
        # Unpaired reads separately
        unpaired1="trimmed/pe/{sample}.u1.fastq",
        unpaired2="trimmed/pe/{sample}.u2.fastq",
        # or in a single file
#        unpaired="trimmed/pe/{sample}.singletons.fastq",
        merged="trimmed/pe/{sample}.merged.fastq",
        failed="trimmed/pe/{sample}.failed.fastq",
        html="report/pe/{sample}.html",
        json="report/pe/{sample}.json"
    log:
        "logs/fastp/pe/{sample}.log"
    params:
        adapters="--adapter_sequence ACGGCTAGCTA --adapter_sequence_r2 AGATCGGAAGAGCACACGTCTGAACTCCAGTCAC",
        extra="--merge"
    threads: 2
    wrapper:
        "v1.25.0-52-g79342b73/bio/fastp"

rule fastp_pe_wo_trimming:
    input:
        sample=["reads/pe/{sample}.1.fastq", "reads/pe/{sample}.2.fastq"]
    output:
        html="report/pe_wo_trimming/{sample}.html",
        json="report/pe_wo_trimming/{sample}.json"
    log:
        "logs/fastp/pe_wo_trimming/{sample}.log"
    params:
        extra=""
    threads: 2
    wrapper:
        "v1.25.0-52-g79342b73/bio/fastp"

Note that input, output and log file paths can be chosen freely.

When running with

snakemake --use-conda

the software dependencies will be automatically deployed into an isolated environment before execution.

Notes

• The adapters param allows to specify adapter sequences
• The extra param allows for additional program arguments.
• For more inforamtion see, https://github.com/OpenGene/fastp

Software dependencies

• fastp=0.23.2

Input/Output

Input:

• fastq file(s)

Output:

• trimmed fastq file(s)
• unpaired reads (optional; eihter in a single file or separate)
• merged reads (optional)
• failed reads (optional)
• json file containing trimming statistics
• html file containing trimming statistics

Authors

• Sebastian Kurscheid (sebastian.kurscheid@unibas.ch)
• Filipe G. Vieira

Code

__author__ = "Sebastian Kurscheid"
__copyright__ = "Copyright 2019, Sebastian Kurscheid"
__email__ = "sebastian.kurscheid@anu.edu.au"
__license__ = "MIT"

from snakemake.shell import shell
import re

extra = snakemake.params.get("extra", "")
adapters = snakemake.params.get("adapters", "")
log = snakemake.log_fmt_shell(stdout=True, stderr=True)


# Assert input
n = len(snakemake.input.sample)
assert (
    n == 1 or n == 2
), "input->sample must have 1 (single-end) or 2 (paired-end) elements."


# Input files
if n == 1:
    reads = "--in1 {}".format(snakemake.input.sample)
else:
    reads = "--in1 {} --in2 {}".format(*snakemake.input.sample)


# Output files
trimmed_paths = snakemake.output.get("trimmed", None)
if trimmed_paths:
    if n == 1:
        trimmed = "--out1 {}".format(snakemake.output.trimmed)
    else:
        trimmed = "--out1 {} --out2 {}".format(*snakemake.output.trimmed)

        # Output unpaired files
        unpaired = snakemake.output.get("unpaired", None)
        if unpaired:
            trimmed += f" --unpaired1 {unpaired} --unpaired2 {unpaired}"
        else:
            unpaired1 = snakemake.output.get("unpaired1", None)
            if unpaired1:
                trimmed += f" --unpaired1 {unpaired1}"
            unpaired2 = snakemake.output.get("unpaired2", None)
            if unpaired2:
                trimmed += f" --unpaired2 {unpaired2}"

        # Output merged PE reads
        merged = snakemake.output.get("merged", None)
        if merged:
            if not re.search(r"--merge\b", extra):
                raise ValueError(
                    "output.merged specified but '--merge' option missing from params.extra"
                )
            trimmed += f" --merged_out {merged}"
else:
    trimmed = ""


# Output failed reads
failed = snakemake.output.get("failed", None)
if failed:
    trimmed += f" --failed_out {failed}"


# Stats
html = "--html {}".format(snakemake.output.html)
json = "--json {}".format(snakemake.output.json)


shell(
    "(fastp --thread {snakemake.threads} "
    "{extra} "
    "{adapters} "
    "{reads} "
    "{trimmed} "
    "{json} "
    "{html} ) {log}"
)