PNC Skill (Dataset-Orchestration Layer)

Overview

pnc-skill is the NeuroClaw orchestration skill for the Philadelphia Neurodevelopmental Cohort (PNC) dataset, a large-scale collaborative study between the University of Pennsylvania and the Children's Hospital of Philadelphia.

It strictly follows the NeuroClaw hierarchical design principles:

• This skill only describes WHAT needs to be done and which tool skill to delegate to.
• It contains no implementation code or concrete commands.
• All concrete execution is delegated to existing base/tool skills via claw-shell.
• Companion scripts in scripts/ provide reference implementations for BIDS validation, phenotype extraction, and QC.

Core workflow (never bypassed):

1. Identify input PNC data and target modalities.
2. Generate a numbered execution plan clearly stating WHAT needs to be done and which tool skill will handle each step.
3. Present the full plan, estimated runtime, resource requirements, and risks to the user and wait for explicit confirmation ("YES" / "execute" / "proceed").
4. On confirmation, delegate every step to the appropriate skill via claw-shell.
5. After execution, save all outputs in a clean directory structure (pnc_output/).

Research use only.

---

Quick Reference

Task	What needs to be done	Delegate to	Expected output
BIDS validation	Validate PNC BIDS structure	scripts/validate_pnc.py	Validation report
sMRI processing	Brain extraction, tissue segmentation	smri-skill	smri_output/ derivatives
rs-fMRI processing	Preprocessing, denoising, connectivity	fmri-skill	fmri_output/ connectivity
task-fMRI processing	Go/No-Go, emotion, memory task GLM	fmri-skill	fmri_output/ task results
dMRI processing	Diffusion preprocessing, tensor metrics	dwi-skill	dwi_output/ metrics
Phenotype extraction	Cognitive, psychiatric, demographic	scripts/extract_pnc_phenotype.py	Merged phenotype CSV
QC summary	Per-subject quality control	scripts/pnc_qc_summary.py	QC summary + exclusion list

---

Dataset Characteristics

• Cohort: ~9,000+ youth aged 8-21 years
• Scanner: 3T Siemens TIM Trio
• Modalities: T1w sMRI, rs-fMRI, task-fMRI, dMRI/DTI
• Task paradigms: Go/No-Go, Fraternal Twins, Penn Line Orientation, Penn Word Memory
• Clinical: Psychiatric assessment, cognitive battery (Penn CNB)
• Access: NIMH Data Archive (NDA), OpenNeuro ds000030 (BIDS subset)
• Format: BIDS-compliant (community conversion)
• Reference: Satterthwaite et al. (2014), NeuroImage

---

Supported Modalities

Modality	Description	Tasks/Conditions
T1w	High-resolution structural MRI	1mm isotropic
rs-fMRI	Resting-state functional MRI	Eyes open
task-fMRI	Task-based functional MRI	Go/No-Go, Emotion, Line Orientation, Word Memory
dMRI	Diffusion-weighted imaging	DTI, white matter tractography

---

PNC Task Paradigms

Task	Description	Cognitive Domain
Go/No-Go	Response inhibition / impulse control	Executive function
Fraternal Twins	Emotion recognition	Social cognition
Penn Line Orientation	Spatial processing	Visuospatial
Penn Word Memory	Memory encoding/retrieval	Episodic memory

---

BIDS Preparation

Script: scripts/validate_pnc.py

Validates PNC BIDS structure and generates a compliance report.

python skills/pnc-skill/scripts/validate_pnc.py \
  --input /path/to/PNC/bids \
  --output /path/to/pnc_output/qc/bids_validation.csv

Features:

• BIDS directory structure validation
• Modality completeness check (T1w, rs-fMRI, task-fMRI, dMRI)
• Age range verification (8-21 years)
• Task paradigm presence check

---

Core Workflow (Never Bypassed)

1. Identify user target: full PNC processing, imaging subset, phenotype extraction, or BIDS validation only.
2. Generate a numbered plan with tools, outputs, runtime, storage, and risks.
3. Wait for explicit confirmation (YES / execute / proceed).
4. On confirmation, run BIDS validation using scripts/validate_pnc.py.
5. Delegate to smri-skill for structural MRI processing.
6. Delegate to fmri-skill for rs-fMRI and task-fMRI processing.
7. Delegate to dwi-skill for dMRI processing.
8. If phenotype extraction is requested, run scripts/extract_pnc_phenotype.py.
9. If QC summary is requested, run scripts/pnc_qc_summary.py.
10. Save outputs into pnc_output/.

---

Modality Processing Delegation

Modality	Delegated skill	Typical tasks	Main outputs
sMRI (T1w)	smri-skill	brain extraction, tissue segmentation	smri_output/ derivatives
rs-fMRI	fmri-skill	preprocessing, denoising, connectivity	fmri_output/ connectivity
task-fMRI	fmri-skill	task GLM, activation analysis	fmri_output/ task results
dMRI	dwi-skill	diffusion preprocessing, tensor metrics	dwi_output/ metrics

---

Standard Output Layout

pnc_output/
├── bids/                   # BIDS-staged data (or validation report)
├── smri/                   # Structural MRI derivatives
├── fmri/                   # Functional MRI derivatives (rs + task)
├── dwi/                    # Diffusion MRI derivatives
├── phenotype/              # Merged phenotype tables (cognitive, psychiatric)
├── qc/                     # QC summaries and exclusion lists
└── logs/                   # Processing logs

---

Benchmark Adapter Guidance

For benchmark-style prompts, do not force the full orchestration when the task only asks for local PNC data validation.

• If the task starts from PNC data already present on disk and only asks for BIDS validation:
  • Skip the download stage
  • Default to the narrow path local PNC discovery -> BIDS validation -> report
• In benchmark mode, do not require explicit confirmation before presenting the validation solution.

---

Safety and Execution Policy

• No execution before explicit plan confirmation.
• All execution must be routed via claw-shell.
• Missing dependencies must be resolved by dependency-planner before running.

---

Important Notes and Limitations

• PNC is a developmental cohort; analyses should account for age effects (8-21 years).
• Pediatric data may require adjusted preprocessing parameters (e.g., higher motion thresholds).
• Penn CNB (Computerized Neurocognitive Battery) provides rich cognitive phenotyping.
• Psychiatric assessment includes DSM-based diagnoses.
• Large sample size enables well-powered developmental analyses.
• pnc-skill is orchestration-only; detailed preprocessing logic remains in modality skills.

---

When to Call This Skill

• User asks for end-to-end PNC workflow.
• User asks to process PNC neuroimaging data.
• User needs BIDS validation for PNC data.
• User asks to extract PNC phenotype data (cognitive, psychiatric, demographic).
• User asks for developmental neuroimaging analysis.

---

Complementary / Related Skills

• smri-skill → structural MRI preprocessing
• fmri-skill → functional MRI preprocessing and analysis
• dwi-skill → diffusion MRI preprocessing
• bids-organizer → BIDS validation and organization
• brain-visualization → visualization of derivatives
• dependency-planner → dependency resolution
• conda-env-manager → environment management
• claw-shell → command execution

---

Reference

• PNC: https://www.med.upenn.edu/bbl/
• Satterthwaite et al. (2014): Neuroimaging of the Philadelphia Neurodevelopmental Cohort. NeuroImage.
• OpenNeuro ds000030
• NIMH Data Archive: https://nda.nih.gov/

Created At: 2026-05-06 13:55 HKT Last Updated At: 2026-05-06 13:55 HKT Author: chengwang96