Skip to content

Copy commands and create sample tables

Examples use a Linux Bash terminal. Follow one tutorial from top to bottom; you do not need to understand Python or write a special run script.

Paths and command lines

  • Replace /path/to/my/analyses_dir/ with an existing folder where you keep analyses.
  • cd changes the current folder; pwd prints it. $PWD inserts its absolute path.
  • mkdir -p creates missing folders. It does not supply the example FASTQ files.
  • Keep quotes around paths, especially if a folder name contains spaces.
  • A final \ continues one command on the next line. Paste the entire block; do not add spaces after the backslash.
  • A flag starts with --. In --config /work/study/config/run.yml, the flag selects which settings file to read.

Create a CSV with cat

CSV means comma-separated values: a header followed by one row per sample. This example creates a real file, not just text displayed in the terminal:

mkdir -p "$PWD/example-input"
cat > "$PWD/example-input/samples.csv" <<'CSV'
sample_name,status
sampleA,TUMOR
sampleB,NORMAL
CSV

Paste everything through the last CSV, which must be on its own line. If the terminal shows > and waits, it is still expecting that closing line; Ctrl+C cancels an unfinished command. cat > replaces an existing file of the same name. Choose a new filename if you need to keep the old table.

<<'CSV' keeps the table text literal. Examples that need $PWD expanded into file paths use <<CSV without quotes. Follow the delimiter shown in each example. Keep the header unchanged. Quote a CSV field containing a comma; use simple, de-identified sample IDs without spaces.

Which table do I need?

Command / purpose Header One row represents
auto --mode illumina sample_name,status One library with matching FASTQ filenames
setup --samplesheet sample,fastq_1,fastq_2,status One library with explicit read paths
auto --mode ont barcode,sample_name,status One sample in one barcode folder
Optional pathology illumina_sample_id,case_code,final_diagnosis One matched, de-identified sample

Do not pass the two-column table to setup --samplesheet. Follow batch setup or explicit FASTQ paths for complete examples.

Check before running

setup and auto save settings. check lists selected samples and catches configuration problems without starting analysis. run starts analysis. Read each result before continuing; if a command fails, fix the error first. Do not paste a run command while the previous command is still running.

A successful check is not a read-quality assessment or a confirmed diagnosis. Review the run summary and QC after analysis.