Add a multiqc report and containerize the trimming

[cokelaer]

Follow-up to #11, whose merge did not include this commit.

MultiQC

A multiqc rule aggregates the FastQC and cutadapt results into
multiqc/multiqc_report.html, which each sample report links from its header.
It uses the multiqc/run shell of sequana-wrappers and the multiqc container,
as the other pipelines do. A multiqc section was added to config.yaml and
to the schema, and multiqc was declared in tools.txt and environment.yml.

Containerized trimming

The trimming used to assemble its command line inside a python run:
directive, which cannot be executed in a container. It is now a plain shell:
rule using the cutadapt container (cutadapt 4.6), with single-end and
paired-end variants, and the declared threads are passed to cutadapt -j.

The atropos variant is kept as a separate rule. No container exists for
atropos, so it must be available locally. Its own json/txt report is no longer
overwritten by the standard output.

Logs

fastq_stat_samples, fastq_stat_phix, fastq_stat_trimmed and
bwa_bam_to_fastq now have a log: file, written as the rule progresses so
that the pipeline monitor reports the real elapsed time instead of a few
seconds. Those four rules stay in python: they call the sequana API, which the
report generation requires on the host anyway. summary is now a local rule.

Also requires sequana >= 0.24.1. Version 1.3.0.

Validation

• dry run of four option combinations: 17, 10, 14 and 7 jobs, no error
• full run without container: exit 0, multiqc report and summary.html created,
  all log files written
• full run with --apptainer-prefix: exit 0, and the log shows the container
  was used (This is cutadapt 4.6 versus cutadapt 5.2 on the host)
• pytest test/test_main.py: 8 passed

🤖 Generated with Claude Code

[codacy-production]

Not up to standards ⛔

_{NEW Get contextual insights on your PRs based on Codacy's metrics, along with PR and Jira context, without leaving GitHub. Enable AI reviewer
TIP This summary will be updated as you push new changes.}

[Copilot]

Pull request overview

Adds MultiQC aggregation and improves portability/observability of the QC pipeline by running cutadapt trimming in a container and writing progress logs for long-running Python rules.

Changes:

• Add a multiqc Snakemake rule producing multiqc/multiqc_report.html, plus config/schema/tooling declarations and a link from sample reports.
• Refactor trimming into containerized shell: cutadapt rules (single/paired) while keeping an atropos fallback rule.
• Add log: files to several Python run: rules and make summary a local rule; bump version to 1.3.0 and require sequana >= 0.24.1.

Reviewed changes

Copilot reviewed 7 out of 7 changed files in this pull request and generated 3 comments.

Show a summary per file

File	Description
sequana_pipelines/quality_control/tools.txt	Declares multiqc as a pipeline tool.
sequana_pipelines/quality_control/schema.yaml	Adds multiqc config validation schema.
sequana_pipelines/quality_control/quality_control.rules	Implements MultiQC rule/linking, containerized cutadapt trimming, and adds progress logs/localrules.
sequana_pipelines/quality_control/config.yaml	Adds cutadapt/multiqc apptainer images and new multiqc config section.
README.rst	Documents new 1.3.0 features/requirements in the changelog.
pyproject.toml	Bumps version, adds keyword, and raises sequana minimum version.
environment.yml	Adds multiqc to the environment dependencies.

---

💡 Add a code-review agent skill or configure MCP servers for context-aware, tailored reviews. Learn more in the docs.

[Copilot]

Pull request overview

Copilot reviewed 8 out of 8 changed files in this pull request and generated no new comments.

Suppressed comments (4)

Previously missed (1) — in code that hasn't changed since the last review.

sequana_pipelines/quality_control/quality_control.rules:106

• multiqc is a new optional config section, but the workflow indexes it with config['multiqc'], which raises a KeyError when running with an older/minimal config that doesn’t define multiqc. Use config.get('multiqc', {}) (or manager.config.multiqc) so the pipeline can still run when the section is absent and MultiQC is effectively disabled.

This issue also appears in the following locations of the same file:

• line 487
• line 494
• line 644

if config['multiqc']['do']:
    expected_output += ["multiqc/multiqc_report.html"]

sequana_pipelines/quality_control/quality_control.rules:488

• This if config['multiqc']['do']: guard assumes the multiqc section always exists. If a config file omits the multiqc block, the workflow fails at parse time with KeyError before any rules run. Consider guarding with config.get('multiqc', {}).get('do', False) so MultiQC is simply skipped when not configured.

if config['multiqc']['do']:

sequana_pipelines/quality_control/quality_control.rules:646

• The MultiQC link injection also assumes config['multiqc'] is present. With an older/minimal config file, report generation will crash with KeyError even if MultiQC isn’t intended to run. Guard this with config.get('multiqc', {}).get('do', False) (or a shared multiqc_enabled flag).

        multiqc_link = ''
        if config['multiqc']['do']:
            multiqc_link = ' | <a href="../../multiqc/multiqc_report.html">MultiQC</a>'

sequana_pipelines/quality_control/quality_control.rules:498

• In the multiqc rule, the schema marks several keys under multiqc as optional (required: False), but the rule accesses them via config['multiqc']['...']. If a user removes one of these optional keys from their config, the workflow will raise a KeyError at parse time. Using .get(..., default) avoids hard failures and keeps the config’s optionality real.

        params:
            options=config['multiqc']['options'],
            input_directory=config['multiqc']['input_directory'],
            config_file=config['multiqc']['config_file'],
            modules=config['multiqc']['modules']