Mock six-tumor/four-normal study¶

This example runs ten paired-end Illumina samples in one analysis. Every sample is aligned, normalized, segmented, called, and reported independently by qDNAseq. OncoTracer does not pool the four CTRL samples or subtract them from tumors. NORMAL records a sample's role, not a reference-building instruction. These example controls are not asserted to be patient-matched normals for somatic variant calling; this workflow requests CNA only.

These 20 FASTQs are not bundled. Use your own de-identified data, changing the names below to match. For checksum-validated public-data examples, choose QuickStart 1 or QuickStart 2.

Terminal / headless version: every step below runs without a browser.

1. Check the installation¶

The automatic sample queue checks available CPU and RAM. For explicit concurrency and total budgets, see multiple samples and continuation.

Follow installation. If you have not installed the Conda tools:

oncotracer install --conda
oncotracer doctor --backend conda

The ordinary OncoTracer command does not invoke Nextflow or a special study script.

2. Create the sample table¶

Replace the analysis directory, then paste the whole block through CSV. cat > replaces the named file if it exists.

cd /path/to/my/analyses_dir/
mkdir -p "$PWD/oncotracer-onco6-ctrl4/input/fastq"
cat > "$PWD/oncotracer-onco6-ctrl4/input/samples.csv" <<'CSV'
sample_name,status
ONCO001,TUMOR
ONCO002,TUMOR
ONCO003,TUMOR
ONCO004,TUMOR
ONCO005,TUMOR
ONCO006,TUMOR
CTRL001,NORMAL
CTRL002,NORMAL
CTRL003,NORMAL
CTRL004,NORMAL
CSV

Place exactly one matching R1/R2 pair per row in input/fastq/:

ONCO001_R1.fastq.gz   ONCO001_R2.fastq.gz
ONCO002_R1.fastq.gz   ONCO002_R2.fastq.gz
ONCO003_R1.fastq.gz   ONCO003_R2.fastq.gz
ONCO004_R1.fastq.gz   ONCO004_R2.fastq.gz
ONCO005_R1.fastq.gz   ONCO005_R2.fastq.gz
ONCO006_R1.fastq.gz   ONCO006_R2.fastq.gz
CTRL001_R1.fastq.gz   CTRL001_R2.fastq.gz
CTRL002_R1.fastq.gz   CTRL002_R2.fastq.gz
CTRL003_R1.fastq.gz   CTRL003_R2.fastq.gz
CTRL004_R1.fastq.gz   CTRL004_R2.fastq.gz

Each name before _R1/_R2 must match the table. Do not merge different libraries. Flag and filename explanations.

3. Save and check the settings¶

cd /path/to/my/analyses_dir/
oncotracer auto \
  --mode illumina \
  --reads-folder "$PWD/oncotracer-onco6-ctrl4/input/fastq" \
  --sample-table "$PWD/oncotracer-onco6-ctrl4/input/samples.csv" \
  --config-dir "$PWD/oncotracer-onco6-ctrl4/config" \
  --outdir "$PWD/oncotracer-onco6-ctrl4/results" \
  --threads 4
oncotracer check --config "$PWD/oncotracer-onco6-ctrl4/config/illumina.auto.yml"

Expect Selected samples: 10 (6 TUMOR, 4 NORMAL) and all ten names in check. Batch setup checks compressed reads and rejects duplicates or missing mates. It saves illumina.auto.yml, illumina.samplesheet.csv and a checksum record, auto_params_manifest.tsv. Keep these together.

To reuse genome indexes, add --hg38_build /path/to/reference to auto. To build your own, add --build_reference instead. With neither flag, run downloads prebuilt indexes. RAM and reference choices.

4. Run or resume the one native analysis¶

cd /path/to/my/analyses_dir/
oncotracer run --backend conda \
  --config "$PWD/oncotracer-onco6-ctrl4/config/illumina.auto.yml"

This analyzes all ten rows. To resume, repeat only this command. Leave the configuration, inputs and output directory unchanged; leave --force off.

5. Confirm that every sample completed¶

cd /path/to/my/analyses_dir/
cat "$PWD/oncotracer-onco6-ctrl4/config/illumina.samplesheet.csv"
cat "$PWD/oncotracer-onco6-ctrl4/results/01_samurai_illumina/qdnaseq/qdnaseq_sample_status.json"
cat "$PWD/oncotracer-onco6-ctrl4/results/06_workflow_summary/workflow_summary.txt"

The samplesheet should retain six tumor and four normal rows. In qdnaseq_sample_status.json, completed_samples must contain all ten names and failed_samples must be empty. The summary records engine=native and nextflow_used=false. Resolve failures before describing the cohort as complete.

Review each sample's plots under 01_samurai_illumina/qdnaseq/plots/ and the final events in 03_cna_codification/cna_events.tsv. A successful run does not confirm a tumor diagnosis. Output guide.