Bioinformatics Pipeline Builder
When to activate
When processing sequencing data (RNA-seq, WGS, ChIP-seq, ATAC-seq, scRNA-seq), building analysis workflows, or when optimizing existing pipelines. Use for any NGS data processing task.
When NOT to use
Skip for simple BLAST searches, single-gene analysis, or when using a validated, unchanged pipeline on routine data.
Instructions
Define the analysis:
- Data type: RNA-seq, WGS, WES, ChIP-seq, ATAC-seq, scRNA-seq, metagenomics
- Organism and genome build (GRCh38, mm39, etc.)
- Sequencing platform and read length
- Research question and expected output
Pipeline stages:
- QC: FastQC, MultiQC, adapter trimming (Trim Galore/fastp)
- Alignment: STAR/HISAT2 (RNA-seq), BWA-MEM2 (DNA-seq), Cell Ranger (scRNA-seq)
- Processing: Sort, index, dedup (Picard/samtools), strand-specific handling
- Quantification: featureCounts/Salmon (RNA-seq), GATK (variants), MACS2 (ChIP-seq)
- Downstream: DESeq2/edgeR (DE), clusterProfiler (GSEA), Seurat/Scanpy (scRNA-seq)
Quality checkpoints:
- Post-QC: Reads passing filter, adapter content, per-base quality
- Post-alignment: Mapping rate (>80%), multi-mapping rate, insert size
- Post-quantification: Gene detection (>1 CPM in ≥X samples), library complexity
Reproducibility requirements:
- Tool versions (exact, not "latest")
- Container images (Docker/Singularity)
- Parameter files (YAML/JSON)
- Random seeds where applicable
Resource estimation:
- RAM, CPU, disk per sample
- Estimated runtime
- Parallelization strategy
Output Format
PIPELINE: [Name] — [Data type]
GENOME: [Build] | ANNOTATION: [GTF/GFF source]
PLATFORM: [Illumina NovaSeq/PacBio/etc.] | READ: [PE150/SE75/etc.]
STAGES:
1. QC: [tool] v[version] — [parameters]
CHECKPOINT: [threshold]
2. Alignment: [tool] v[version] — [parameters]
CHECKPOINT: Mapping rate ≥[X]%
3. [Processing steps]
4. Quantification: [tool] v[version]
CHECKPOINT: [threshold]
5. Downstream: [tool/method]
RESOURCES:
Per sample: [X] GB RAM, [Y] CPU, [Z] GB disk
Estimated total: [time]
REPRODUCIBILITY:
Container: [Docker image:tag]
Parameters: [config file path]
Seed: [value where applicable]
Example
PIPELINE: Differential Gene Expression — Bulk RNA-seq
GENOME: GRCh38 (GENCODE v44) | ANNOTATION: gencode.v44.annotation.gtf
PLATFORM: Illumina NovaSeq 6000 | READ: PE150
STAGES:
1. QC: FastQC v0.12.1 + MultiQC v1.14
CHECKPOINT: Per-base quality ≥Q30, adapter content <5%
2. Trimming: Trim Galore v0.6.10 --paired --quality 20 --length 50
3. Alignment: STAR v2.7.11a --outSAMstrandField intronMotif --quantMode GeneCounts
CHECKPOINT: Mapping rate ≥85%, uniquely mapped ≥70%
4. Quantification: Salmon v1.10.1 (alignment-based mode)
CHECKPOINT: ≥10,000 genes detected (>1 CPM in ≥3 samples)
5. DE Analysis: DESeq2 v1.42.0, Wald test, BH correction
CHECKPOINT: |log2FC| ≥ 1, padj < 0.05
6. Enrichment: clusterProfiler v4.10.0, GO + KEGG
RESOURCES:
Per sample: 32 GB RAM, 8 CPU, 50 GB disk
Estimated total: ~4h for 24 samples
REPRODUCIBILITY:
Container: biocontainers/star:2.7.11a
Parameters: config/rnaseq-pipeline.yaml
1---2name: bioinformatics-pipeline-builder3description: Designs and documents bioinformatics analysis pipelines for genomics, transcriptomics, and proteomics data. Outputs reproducible pipeline definitions with tool versions, parameters, and quality control checkpoints.4---56# Bioinformatics Pipeline Builder78## When to activate9When processing sequencing data (RNA-seq, WGS, ChIP-seq, ATAC-seq, scRNA-seq), building analysis workflows, or when optimizing existing pipelines. Use for any NGS data processing task.1011## When NOT to use12Skip for simple BLAST searches, single-gene analysis, or when using a validated, unchanged pipeline on routine data.1314## Instructions15161. **Define the analysis:**17 - Data type: RNA-seq, WGS, WES, ChIP-seq, ATAC-seq, scRNA-seq, metagenomics18 - Organism and genome build (GRCh38, mm39, etc.)19 - Sequencing platform and read length20 - Research question and expected output21222. **Pipeline stages:**23 - **QC:** FastQC, MultiQC, adapter trimming (Trim Galore/fastp)24 - **Alignment:** STAR/HISAT2 (RNA-seq), BWA-MEM2 (DNA-seq), Cell Ranger (scRNA-seq)25 - **Processing:** Sort, index, dedup (Picard/samtools), strand-specific handling26 - **Quantification:** featureCounts/Salmon (RNA-seq), GATK (variants), MACS2 (ChIP-seq)27 - **Downstream:** DESeq2/edgeR (DE), clusterProfiler (GSEA), Seurat/Scanpy (scRNA-seq)28293. **Quality checkpoints:**30 - Post-QC: Reads passing filter, adapter content, per-base quality31 - Post-alignment: Mapping rate (>80%), multi-mapping rate, insert size32 - Post-quantification: Gene detection (>1 CPM in ≥X samples), library complexity33344. **Reproducibility requirements:**35 - Tool versions (exact, not "latest")36 - Container images (Docker/Singularity)37 - Parameter files (YAML/JSON)38 - Random seeds where applicable39405. **Resource estimation:**41 - RAM, CPU, disk per sample42 - Estimated runtime43 - Parallelization strategy4445## Output Format4647```48PIPELINE: [Name] — [Data type]49GENOME: [Build] | ANNOTATION: [GTF/GFF source]50PLATFORM: [Illumina NovaSeq/PacBio/etc.] | READ: [PE150/SE75/etc.]5152STAGES:53 1. QC: [tool] v[version] — [parameters]54 CHECKPOINT: [threshold]55 2. Alignment: [tool] v[version] — [parameters]56 CHECKPOINT: Mapping rate ≥[X]%57 3. [Processing steps]58 4. Quantification: [tool] v[version]59 CHECKPOINT: [threshold]60 5. Downstream: [tool/method]6162RESOURCES:63 Per sample: [X] GB RAM, [Y] CPU, [Z] GB disk64 Estimated total: [time]6566REPRODUCIBILITY:67 Container: [Docker image:tag]68 Parameters: [config file path]69 Seed: [value where applicable]70```7172## Example7374```75PIPELINE: Differential Gene Expression — Bulk RNA-seq76GENOME: GRCh38 (GENCODE v44) | ANNOTATION: gencode.v44.annotation.gtf77PLATFORM: Illumina NovaSeq 6000 | READ: PE1507879STAGES:80 1. QC: FastQC v0.12.1 + MultiQC v1.1481 CHECKPOINT: Per-base quality ≥Q30, adapter content <5%82 2. Trimming: Trim Galore v0.6.10 --paired --quality 20 --length 5083 3. Alignment: STAR v2.7.11a --outSAMstrandField intronMotif --quantMode GeneCounts84 CHECKPOINT: Mapping rate ≥85%, uniquely mapped ≥70%85 4. Quantification: Salmon v1.10.1 (alignment-based mode)86 CHECKPOINT: ≥10,000 genes detected (>1 CPM in ≥3 samples)87 5. DE Analysis: DESeq2 v1.42.0, Wald test, BH correction88 CHECKPOINT: |log2FC| ≥ 1, padj < 0.0589 6. Enrichment: clusterProfiler v4.10.0, GO + KEGG9091RESOURCES:92 Per sample: 32 GB RAM, 8 CPU, 50 GB disk93 Estimated total: ~4h for 24 samples9495REPRODUCIBILITY:96 Container: biocontainers/star:2.7.11a97 Parameters: config/rnaseq-pipeline.yaml98```
UitbreidenOS