skills.rest
CUHK-AIM-Group
CUHK-AIM-GroupOFFICIAL
Explore all skills by this author →
WebsiteGitHub

abide-skill

Orchestrates ABIDE dataset download, BIDS staging, phenotype extraction, and multimodal neuroimaging processing.

89|5|Updated Mar 25, 2026
Search Tags:#quality control#autism#neuroimaging#fmri#bids#phenotype extraction#abide
One-click install
npx skills add https://github.com/CUHK-AIM-Group/NeuroDiscovery --skill abide-skill-cuhk-aim-group
Or copy as Structured Prompt for Agent▼
Please help me install this Agent Skill.
Skill: abide-skill
Source: https://github.com/CUHK-AIM-Group/NeuroDiscovery/tree/main/skills/abide-skill
Command: npx skills add https://github.com/CUHK-AIM-Group/NeuroDiscovery --skill abide-skill-cuhk-aim-group

SYSTEM DOCUMENTATION & REQUIREMENTS

💡 This Skill requires pandas, and includes scripts (resource) components.

What problem does it solve? Working with the ABIDE (Autism Brain Imaging Data Exchange) dataset requires coordinating downloads from FCP/INDI or NITRC, reorganizing inconsistent raw NIfTI layouts into BIDS-compliant structure, merging phenotype tables, and running sMRI and rs-fMRI pipelines. This Skill provides a single orchestration layer that manages that entire workflow. ## Core Features & Use Cases - BIDS Reorganization: Converts site-based or flat ABIDE raw NIfTI layouts into BIDS-compliant datasets with subject ID normalization, site encoding, modality routing (T1w, rs-fMRI), sidecar preservation, and dataset_description.json/participants.tsv generation. - Phenotype Extraction: Merges ABIDE I and II phenotype CSVs, standardizes diagnosis labels (ASD/control), selects columns, and cross-references with imaging subject lists. - QC Summaries: Combines fMRIPrep confounds (framewise displacement, DVARS), FreeSurfer recon-all metrics, and ABIDE native QC flags into per-subject QC reports with configurable exclusion thresholds. - Use Case: A researcher downloads ABIDE I raw data, stages it into BIDS format, merges phenotype tables for 1,112 subjects, then delegates structural and functional processing to modality-specific pipelines with a unified QC exclusion list. ## Quick Start Ask the agent to download ABIDE I data, reorganize it into a BIDS dataset under abide_output/bids, and extract the merged phenotype table for ASD versus control analysis.

Frequently Asked Questions about abide-skill

High-intent search queries and answers about installing and using this skill.

FAQPage Schema
How do I convert ABIDE raw data to BIDS format?▼

Run the reorganize_abide.py script with --input pointing to the raw ABIDE directory and --output for the BIDS target. It auto-detects site-based or flat layouts, normalizes subject IDs to sub-NNNNNNN labels, routes T1w and rs-fMRI files, and writes dataset_description.json and participants.tsv. Use --dry-run to preview first.

How to merge ABIDE phenotype tables with imaging subject lists?▼

Use extract_abide_phenotype.py with --phenotype-dir or --phenotype-file for the CSV source and --imaging-ids pointing to a BIDS participants.tsv. It standardizes column names, maps DX_GROUP values to ASD/control labels, and filters rows to subjects with imaging data.

Does ABIDE include task-based fMRI data?▼

No, ABIDE only provides resting-state fMRI alongside structural T1w scans. Task-fMRI is not available in either ABIDE I or ABIDE II, so analyses are limited to resting-state connectivity and structural morphometry.

What QC thresholds are used to exclude ABIDE subjects?▼

The QC script applies a default mean framewise displacement threshold of 0.3 mm and max FD of 5.0 mm from fMRIPrep confounds, plus FreeSurfer recon-all completion checks. Thresholds are configurable via --fd-threshold, --max-fd-threshold, and --coverage-threshold arguments.

Can I process only ABIDE BIDS staging without full pipeline execution?▼

Yes, the skill supports a narrow path for local raw NIfTI to BIDS staging only. It skips the download stage and modality delegation, producing just the staged dataset, minimal metadata, and a validation report under abide_output/bids.

Related Skills

Explore More Education & Research Skills→
NVIDIANVIDIA
3.2k

i4h-catheter-navigation-digital-twin

Builds a patient vasculature digital twin from CT volumes via preprocessing and vessel segmentation.

Official
Intermediate
NeuroAIHubNeuroAIHub
269

fmriprep

Execute containerized BIDS-App fMRI preprocessing with motion correction and anatomical registration.

Official
Advanced
CUHK-AIM-GroupCUHK-AIM-Group
78

fmriprep-tool

Generate and execute fMRIPrep preprocessing plans for BIDS datasets.

Official
Advanced
CUHK-AIM-GroupCUHK-AIM-Group
78

hcp-skill

Orchestrate HCP multimodal processing from data download to modality-specific pipelines.

Official
Advanced