Module 1: Read trimming¶

Module 1 trims reads that have been discovered and validated by Module 0: Read processing. It runs fastp on paired-end and interleaved reads, runs FastQC on the trimmed reads, and writes per-sample and combined trimming statistics for downstream modules.

Prerequisite¶

Run Module 0 first. By default, Module 1 expects the Module 0 manifest at:

<working_dir>/module_0_readprocess/naming/read_manifest.tsv

If --working_dir is omitted, the resolved SAMWISE source directory is used as the results root. Use --input_manifest to provide a different manifest:

nextflow run module_1_readtrimming.nf \
  --working_dir ./output_samwise \
  --input_manifest ./somewhere/read_manifest.tsv

The manifest must be tab-separated and include the fields used by Module 1: sample_id, layout, and the corresponding read1/read2 fields for paired rows or interleaved for interleaved rows. The referenced input files must exist. Module 0 is responsible for read naming, pairing/layout detection, and optional FASTQ structure validation.

Usage¶

nextflow run module_1_readtrimming.nf \
  --working_dir ./output_samwise \
  --threads 6

The default manifest discovery makes additional input flags unnecessary when running after Module 0 in the same results directory.

Parameters¶

Tool setup and environment¶

The setup process writes setup/module1_tools_status.env and selects tools in this order:

1. Reuse tool_env_dir, or the default managed environment, if it exists and both executables pass --version checks.
2. Otherwise use working fastp and fastqc executables already on PATH.
3. Otherwise, when auto_install=true, create the managed environment with mamba if available, or conda if available, using the conda-forge and bioconda channels.

An existing incomplete or broken managed environment is removed before reinstallation. With auto_install=false, the workflow fails if either tool is missing or unusable. Trimming and FastQC tasks read the selected tool paths from the status file and add the managed environment to PATH when needed.

Output layout¶

For <outdir> = <results_dir>/module_1_readtrimming, published outputs are:

module_1_readtrimming/
├── conda_envs/
│   └── module1_tools/                 # default managed environment, if created
├── setup/
│   └── module1_tools_status.env
├── trimmed_reads/
│   ├── <safe_id>_R1_trimmed.fastq.gz
│   ├── <safe_id>_R2_trimmed.fastq.gz
│   └── <safe_id>_interleaved_trimmed.fastq.gz
├── fastp_reports/
│   ├── <safe_id>_fastp.html
│   ├── <safe_id>_fastp.json
│   └── <safe_id>_fastp.log
├── fastqc_reports/
│   ├── <read>_fastqc.html
│   └── <read>_fastqc.zip
└── summary/
    ├── <safe_id>_trimming_stats.tsv
    ├── trimming_stats_summary.tsv
    └── trimmed_manifest.tsv

trimmed_reads uses publish_trimmed_mode; setup, FastQC, fastp reports, and summary files are published with copy mode. FastQC files are absent when skip_fastqc=true.

summary/trimmed_manifest.tsv has these columns:

sample_id  safe_sample_id  layout  read1  read2  interleaved  merged  fastp_html  fastp_json

The read and report paths in this manifest are absolute published paths so the manifest can be consumed from later Nextflow task directories. Paired rows populate read1 and read2; interleaved rows populate interleaved.

Each per-sample stats file and trimming_stats_summary.tsv report the fastp before/after counts, reads removed by filters, adapter-trimmed reads, final FASTQ record counts, and the associated fastp JSON path. Paired final counts are measured separately for read 1 and read 2; interleaved counts are measured from the single interleaved output.

Validation notes¶

• Module 1 checks that the resolved manifest exists before processing it and fails when no trimmed-manifest records or trimming-stat files are produced.
• The manifest layout value must be exactly paired or interleaved; each layout requires its corresponding path fields.
• Module 0 should be used to validate FASTQ structure. Module 1 additionally counts output FASTQ records and fails if a FASTQ line count is not divisible by four.
• Sample identifiers are converted to safe output names by replacing any character outside [A-Za-z0-9._-] with _. Different identifiers can therefore map to the same safe name; use unique, filesystem-safe sample IDs.
• Check setup/module1_tools_status.env and the FastQC/fastp reports when diagnosing tool or trimming failures.

Next step: Module 2 assembly¶

After a successful run, Module 2 and Module 2B consume summary/trimmed_manifest.tsv from this output directory. The manifest is the handoff between read trimming and assembly; it contains the trimmed read paths, layout, sample IDs, and associated report paths.

Continue with Module 2: Assembly, or use Module 2B: Grouped coassembly when assigning samples to coassembly groups. Both workflows accept --input_manifest when the default trimmed manifest is not being used.