bio-reporting-automated-qc-reports - SKILL.md Agent Skill

name: bio-reporting-automated-qc-reports description: Generates standardized quality control reports by aggregating metrics from FastQC, alignment, and other tools using MultiQC. Use when summarizing QC metrics across samples, creating shareable quality reports, or building automated QC pipelines. tool_type: cli primary_tool: multiqc

Version Compatibility

Reference examples tested with: Cell Ranger 8.0+, FastQC 0.12+, GATK 4.5+, HISAT2 2.2.1+, MultiQC 1.21+, STAR 2.7.11+, Subread 2.0+, bcftools 1.19+, fastp 0.23+, kallisto 0.50+

Before using code patterns, verify installed versions match. If versions differ:

Python: pip show <package> then help(module.function) to check signatures
CLI: <tool> --version then <tool> --help to confirm flags

If code throws ImportError, AttributeError, or TypeError, introspect the installed package and adapt the example to match the actual API rather than retrying.

Automated QC Reports with MultiQC

"Aggregate QC results into one report" → Combine outputs from FastQC, samtools, Picard, and other tools into a single interactive HTML report.

CLI: multiqc . (scans current directory for recognized tool outputs)

Basic Usage

# Aggregate all QC outputs in directory
multiqc results/ -o qc_report/

# Specify output name
multiqc results/ -n my_project_qc

# Include specific tools only
multiqc results/ --module fastqc --module star

Supported Tools

MultiQC recognizes outputs from 100+ bioinformatics tools:

Category	Tools
Read QC	FastQC, fastp, Cutadapt
Alignment	STAR, HISAT2, BWA, Bowtie2
Quantification	featureCounts, Salmon, kallisto
Variant Calling	bcftools, GATK
Single-cell	CellRanger, STARsolo

Configuration

Create multiqc_config.yaml:

title: "RNA-seq QC Report"
subtitle: "Project XYZ"
intro_text: "QC metrics for all samples"

# Custom sample name cleaning
extra_fn_clean_exts:
  - '.sorted'
  - '.dedup'

# Report sections to include
module_order:
  - fastqc
  - star
  - featurecounts

# Highlight samples
table_cond_formatting_rules:
  pct_mapped:
    fail: [{lt: 50}]
    warn: [{lt: 70}]

Custom Data

# Add custom data file
# File format: sample\tmetric1\tmetric2
multiqc results/ --data-format tsv --custom-data-file custom_metrics.tsv

Python API

from multiqc import run as multiqc_run

# Run programmatically
multiqc_run(analysis_dir='results/', outdir='qc_report/')

Related Skills

read-qc/quality-reports - Generate input FastQC reports
read-qc/fastp-workflow - Preprocessing QC
workflows/rnaseq-to-de - Full workflow with QC