Add consolidated D4D datasheet for CM4AI

[d4dassistant]

Summary

Created new consolidated D4D datasheet for Cell Maps for Artificial Intelligence (CM4AI) based on comprehensive documentation from CM4AI website, Virginia Dataverse releases, and 38+ publications.

Files Added

• data/sheets_d4dassistant/cm4ai_d4d.yaml - Consolidated CM4AI D4D datasheet

Validation

• ✅ Schema validation passed
• ✅ Required fields populated (id, name)
• ✅ YAML syntax valid
• ✅ Structured with resources section for detailed component metadata

Key Metadata Extracted

Dataset ID: cm4ai
Dataset Name: Cell Maps for Artificial Intelligence (CM4AI)

Purpose: CM4AI was created to generate comprehensive, AI-ready maps of human cell architecture from disease-relevant cell lines to support interpretable genotype-phenotype learning and advance functional genomics research using the FAIRSCAPE framework and RO-Crate format.

Composition:

• 53,788 immunofluorescence images
• 1,792 proteins investigated
• 11,739 genes targeted via CRISPR
• 1,374 protein interactions mapped
• 22.7 TB total data volume

Data Types:

1. CRISPR Perturbation Cell Atlas: Genome-scale CRISPRi perturbation atlas in KOLF2.1J hiPSCs with single-cell RNA sequencing
2. Protein-Protein Interactions: SEC-MS data from multiple cell types (hiPSCs, NPCs, neurons, cardiomyocytes)
3. Protein Localization Imaging: Multi-channel confocal microscopy of 563 proteins in MDA-MB-468 cells under various treatment conditions

Distribution:

• Repository: University of Virginia Dataverse (https://dataverse.lib.virginia.edu/dataverse/CM4AI)
• License: CC BY-NC-SA 4.0
• Releases: Quarterly updates (Alpha May 2024, Beta March 2025, Beta June 2025)
• Formats: RO-Crate packages, ZIP archives, FASTQ, CSV, H5AD

Consortium: UC San Diego (lead), UCSF, Stanford, UVA, Yale, UT Austin, UAB, Simon Fraser, Hastings Center

Funding: NIH Bridge2AI grant 1OT2OD032742-01

Sources

This consolidated datasheet synthesizes information from:

1. CM4AI Website: https://cm4ai.org (main project page, data releases, tools, publications)
2. Virginia Dataverse Releases:
   • March 2025 Beta: doi:10.18130/V3/B35XWX
   • June 2025 Beta: doi:10.18130/V3/F3TD5R
   • Alpha v0.5: doi:10.18130/V3/DXWOS5
3. CM4AI Publications (https://cm4ai.org/publications/):
   • Clark T, et al. Cell Maps for Artificial Intelligence (bioRxiv 2024.05.21.589311)
   • Schaffer LV, et al. Multimodal cell maps as a foundation for structural and functional genomics (Nature 2025)
   • Lenkiewicz J, et al. Cell Mapping Toolkit (Bioinformatics 2024)
   • Al Manir S, et al. FAIRSCAPE: An Evolving AI-readiness Framework (bioRxiv 2024)
   • Ethics and ELSI publications from Pacia DM, Stevens I, Ravitsky V, et al.
4. Existing Datasheets: Consolidated information from data/extracted_by_column/CM4AI/dataverse_10.18130_V3_B35XWX_d4d.yaml and related files

How to Review

1. Check YAML structure: Review data/sheets_d4dassistant/cm4ai_d4d.yaml for completeness and accuracy
2. Verify sources: Compare against original CM4AI documentation URLs listed above
3. Validate resources section: Three detailed resource entries cover CRISPR atlas, SEC-MS data, and imaging data
4. Check metadata fields: Confirm populated fields are accurate; omitted fields indicate missing source info

Notes

• Fields marked as null or omitted indicate information not found in source documentation
• Controlled vocabulary fields use enums defined in the D4D schema
• All dates follow ISO 8601 format (YYYY-MM-DD)
• Resources section provides detailed metadata for each major data component
• This consolidates information across multiple data releases into a single comprehensive datasheet

Related to: #71

---

🤖 Generated with D4D Assistant