ChromTransfer: Genomic Data for Chromatin State Prediction Models

13.0 GB of genomic data files support preprocessing, training, and prediction for the ChromTransfer model. The collection includes TF co-binding signals, chromatin context, DNA sequences, and FUNCODE scores for mouse (mm10) and human (hg38) genomes. Yiman Wang published this dataset on figshare in April 2026.

Use Cases

• Training ChromTransfer-Base models based on DNA sequences for genomic regions.
• Training ChromTransfer-Cons models based on FUNCODE scores for genomic regions.
• Training ChromTransfer-Reg models based on TF co-binding and chromatin context signals.
• Identifying co-binding transcription factors based on TF interaction information.
• Generating training, validation, and test datasets based on BED files defining genomic regions.

Strengths

• Dataset size is 13.0 GB, indicating a substantial collection of genomic data.
• Data is provided for two major reference genomes: mouse (mm10) and human (hg38).
• Files are clearly organized by function (Regulatory, FUNCODE, DNA, regions, cobinding_TF_source, demo).

Limitations

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

Provenance

Source

figshare

Collection Method

Likely compiled for the ChromTransfer model project.

Freshness

Last updated 2026-04-10 07:11:42; freshness should be verified.