Usage

Tip

Use -profile test for smoke checks before full-scale datasets.

Basic invocation

nextflow run main.nf \
  -profile standard \
  --input data/input.fasta \
  --outdir results

Recommended production invocation (SLURM)

nextflow run main.nf \
  -profile hpc \
  --input /path/to/input.fasta \
  --outdir /path/to/output \
  --max_cpus 96 \
  --max_memory 1000.GB \
  --max_time 240.h

Warning

The pipeline splits records first, so very high header counts (close to 1000) can generate many short jobs and increase scheduler overhead.

Stage graph

flowchart TD
    U1[Run command] --> U2[SPLIT_FASTA]
    U2 --> U3[Per-record fanout]
    U3 --> U4[CODON_ANALYSIS / LONG_ORF / TRANSLATE_FASTA]
    U4 --> U5[HYDROPATHY_PROFILE]
    U5 --> U6[PLOT_HYDROPATHY]
    U4 --> U7[GATHER_RESULTS]
    U5 --> U7
    U6 --> U7