CFM-GP: Single-Cell Gene Perturbation Expression Data

A dataset associated with the CFM-GP deep learning framework for modeling gene perturbation effects across cell types. The data is provided as a 46.8 MB ZIP file by author Sajib Acharjee Dip and was last updated on April 24, 2026. It likely contains gene expression measurements from control and perturbed single-cell experiments.

Use Cases

• Training a unified model to predict gene perturbation effects across diverse cell types based on the described conditional flow matching framework.
• Benchmarking computational methods for single-cell perturbation response prediction against the CFM-GP model's data.
• Analyzing the continuous transformation between control and perturbed gene expression distributions conditioned on cell type.

Strengths

• Dataset is 46.8 MB in size, indicating a manageable download and storage footprint.
• Released under the permissive CC0-1.0 license, allowing for broad reuse and distribution.
• Associated with a published methodological framework (CFM-GP) for unified modeling across cell types.

Limitations

• Row count is unknown, which may limit suitability assessment for large-scale modeling.
• Column-level documentation is absent; field semantics must be inferred after download.
• Description metadata is limited; actual data quality requires manual inspection after download.

Provenance

Source

figshare

Collection Method

Likely derived from single-cell RNA sequencing experiments measuring transcriptional responses to gene perturbations, as described in the associated research.

Freshness

Last updated 2026-04-24 17:53:42; freshness should be verified.