Preparation (prep)

Preparation Tasks¶

CLI flag: --prep [unzip | recon | unzip_recon]

unzip¶

What it does: Extracts all compressed archives (.zip, .7z, .tar.gz) from the input directory using parallel decompression. After extraction it checks for {prefix}{subject_id} directories and prepares them for downstream processing.

Tool: p7zip + GNU parallel
SLURM profile: standard (32 GB, 20 h) — non-array job (all subjects in one job)
Script: unzip_rename.sh
Input: Any directory containing compressed data files
Output: {output}/raw/

Config entry (in {project}_config.yaml):

tasks:
  unzip:
    environ: ["data_manage_1"]    # loads p7zip and parallel modules

recon¶

What it does: Converts DICOM files to BIDS format using dcm2bids inside a Singularity container.

Tool: dcm2bids (Singularity)
SLURM profile: light_short (16 GB, 4 h) — array job
Script: dcm2bids_convert_BIDS.sh
Depends on: unzip
Input: {output}/raw/{subject}/
Output: {output}/BIDS/sub-{subject}/ses-{session}/

Config entry (in {project}_config.yaml):

tasks:
  recon:
    container: "dcm2bids_3.2.0.sif"   # must be in container_dir
    config: "branch_config.json"        # dcm2bids BIDS config file

The config file maps DICOM series descriptions to BIDS modality labels. Store it in config_dir.

Common Usage¶

# Unzip and convert in one step
neuropipe run \
  --subjects 001,002,003 \
  --input /data/zip_files \
  --output /data/processed \
  --work /data/work \
  --project my_study \
  --session 01 \
  --prep unzip_recon

# Unzip only
neuropipe run ... --prep unzip

# Convert already-unzipped data
neuropipe run ... --prep recon