Directory Layout

The repository

PoolSeqFlow/
├── bin/
│   ├── atomic_mv.sh              # Cross-filesystem moves, staged and renamed
│   ├── config_migrate.sh         # Backs migrate_config
│   ├── createDepthFile.sh        # Extract AD/DP columns from a VCF
│   ├── depth2freq.awk            # Convert allelic depths to frequencies
│   ├── filterFalsePositives.sh   # Cross-sample support filter
│   └── MajorAlleleToRef.py       # Re-encode VCF with the major allele as REF
├── install/
│   ├── environment.yml           # Pinned conda environment
│   └── check_install.sh          # Verifies an installation (./PoolSeqFlow check)
├── scripts/
│   ├── 0_verify_environment.nf
│   ├── 1_build_dictionaries.nf
│   ├── 2_trim_reads.nf
│   ├── 3_align.nf
│   ├── 4_clean.nf
│   ├── 5_reports.nf
│   ├── 6_variant_call.nf
│   ├── 7_vcf2freq.nf
│   └── 8_annotate_variants.nf
├── docs/                         # This documentation
├── mkdocs.yml
├── nextflow.config
├── parameters.config             # Yours — not tracked in git
├── parameters.config.template
├── poolseqflow.nf                # Workflow entry point
├── PoolSeqFlow                   # CLI wrapper
└── RGTags.csv.template

bin/ is prepended to PATH by nextflow.config, which is how the helper scripts are callable by bare name inside process scripts.

What you provide

/path/to/project/            ← projectDir
├── Data/                    ← dataSource
│   ├── Sample1_R1.fq.gz
│   ├── Sample1_R2.fq.gz
│   └── …
├── RGTags.csv
├── reference.fasta.gz
└── reference.gff.gz         ← only if annotate = true

Both reference files must be gzipped. The pipeline decompresses them into Reference/ itself.

What the pipeline produces

/path/to/project/
├── .poolseqflow_params           # Analysis parameters behind these outputs
├── .poolseqflow_rgtags           # RGTags.csv as consumed
├── .poolseqflow_versions         # Pipeline versions that have run here, oldest first
├── Logs/                         # Per-step .log and .err, mirrored from every task
├── Reference/
│   ├── reference.fasta
│   ├── reference.fasta.{amb,ann,bwt,fai,pac,sa}
│   └── snpEff/
└── Output/
    ├── run_parameters.txt        # Read-only mirror of .poolseqflow_params
    ├── Trimmed/<sample>/         # Trimmed and clipped FASTQs
    ├── Unpaired/<sample>/        # Reads whose mate was discarded
    ├── Aligned/                  # Raw BWA output
    ├── Ready/                    # Cleaned, filtered, indexed BAMs
    ├── VCF/                      # Call sets
    ├── Frequencies/              # The result
    └── Reports/
        ├── Alignment/
        ├── Coverage/
        ├── Fastqc/<sample>/
        ├── Trimming/<sample>/
        ├── 0_verify_environment.txt
        ├── snpeff_summary.html
        └── PoolSeqFlow_pipeline_{report,timeline,trace,dag}.*

mainDir additionally holds work/, which cleanup = true empties after a successful run and ./PoolSeqFlow clean removes.

What survives a completed run

Several steps delete their inputs once the next stage has consumed them, and the deletion follows the symlink — the permanent copy goes too (why). So the contents of Output/VCF/ mid-run and after a completed run are not the same.

With vcf.fileName = 'Test':

File	Survives	Notes
Test.vcf	Yes	Raw call set from step 6
Test_sort.vcf	No	Deleted by the false-positive filter
Test_sort_fp.vcf	No	Deleted by the depth/quality filter
Test_sort_fp_dq.vcf	Yes	Fully filtered, major-allele normalised
Test_sort_fp_dq_snp.vcf	No	Deleted by frequency conversion
Test_sort_fp_dq_indel.vcf	No	Deleted by frequency conversion
Test_annotated.vcf	Yes	If annotate = true; annotates the raw call set
Test_snp_freq.tsv	Yes	In Frequencies/
Test_indel_freq.tsv	Yes	In Frequencies/

Likewise, Output/Trimmed/<sample>/ keeps only the _clipped.fq.gz files after a completed run — the intermediate _val_1/_val_2 files are removed once clipping has used them.

Output/Aligned/ and Output/Ready/ are both kept. Nothing deletes a BAM.

Sizing storage

Rough guidance for planning, per sample:

Directory	Relative size	Kept?
Data/	Your input	Yours
Trimmed/	Slightly under input, after clipping	Yes
Unpaired/	Small	Yes
Aligned/	Comparable to trimmed input	Yes
Ready/	Smaller — duplicates and filtered reads removed	Yes
VCF/	Depends on variant density, not read count	Partly
Frequencies/	Larger than the VCF it came from — one row per allele, not per site	Yes

The two BAM directories dominate. If disk is tight, Output/Aligned/ is the safe thing to remove after a completed run: step 4 has consumed it, and only a re-run from alignment would need it back.

mainDir needs far less — the symlink strategy means work/ holds links rather than copies, with genuine working space needed only for the intermediates of currently running tasks.