---
title: ReporterNet
canonical_url: "https://www.modelscope.cn/datasets/User040218/ReporterNet"
md_url: "https://www.modelscope.cn/datasets/User040218/ReporterNet.md"
repository: User040218/ReporterNet
last_updated: 2026-11-01
license: other
storage_size: "2.1 GB"
downloads: 1166
stars: 0
---

# ReporterNet

> ReporterNet - User040218 在 ModelScope 开源的数据集。ENCODE ReporterNet training data: MPRA, ATAC-STARR, WG-STARR and lentiMPRA reporter activity (sequence regression).

User040218/ReporterNet 是 ModelScope 魔搭社区上的数据集，存储大小 2.1 GB，采用 other 许可。

- **Repository**: User040218/ReporterNet
- **License**: other
- **Storage size**: 2.1 GB
- **Downloads**: 1166
- **Stars**: 0
- **Last updated**: 2026-11-01

Source: https://www.modelscope.cn/datasets/User040218/ReporterNet

---

# ReporterNet

**Observed training data of the ENCODE ReporterNet models: MPRA, ATAC-STARR, WG-STARR and lentiMPRA reporter activity** — regression.

ReporterNet is a family of ENCODE models that predict massively parallel reporter assay activity, log2(RNA/DNA), from DNA sequence. According to ENCODE, each model's observed training data are included in its model archive; this release extracts the training data of all 8 ReporterNet models (model weights excluded).

Official dataset description: [ENCODE BPNet-family model file index](https://github.com/kundajelab/ENCODE_atlas_bpnet_model_family_files). See also: [ENCSR204SXE](https://www.encodeproject.org/annotations/ENCSR204SXE/), [ENCSR375XWZ](https://www.encodeproject.org/annotations/ENCSR375XWZ/).

## Subsets (10)

| Subset | Splits (rows) | Description |
|---|---|---|
| `MPRA_fold_0` | train 550,877 · validation 88,715 · test 158,472 | MPRA (Gosai et al.), fold 0 |
| `MPRA_fold_1` | train 510,977 · validation 120,832 · test 166,255 | MPRA, fold 1 |
| `MPRA_fold_2` | train 564,081 · validation 87,463 · test 146,520 | MPRA, fold 2 |
| `MPRA_fold_3` | train 562,163 · validation 77,935 · test 157,966 | MPRA, fold 3 |
| `MPRA_fold_4` | train 551,519 · validation 77,694 · test 168,851 | MPRA, fold 4 |
| `ATAC-STARR_K562_ENCSR204SXE` | train 356,843 | K562 ATAC-STARR |
| `WG-STARR_K562_ENCSR375XWZ` | train 769,459 | K562 whole-genome STARR |
| `lentiMPRA_HepG2_ENCSR290IZP` | train 245,852 | lentiMPRA (Agarwal et al.), HepG2 |
| `lentiMPRA_K562_ENCSR149OFN` | train 393,328 | lentiMPRA, K562 |
| `lentiMPRA_WTC11_ENCSR095JXI` | train 92,250 | lentiMPRA, WTC11 |

5,848,052 rows in total.

## Format

Parquet. Main columns:

| Column | Type | Description |
|---|---|---|
| `sequence` | string | DNA sequence (`seq` in the lentiMPRA subsets) |
| `K562_log2FC_float` / `HepG2_log2FC_float` / `SKNSH_log2FC_float` | float64 | MPRA activity per cell type |
| `*_lfcSE_float` | — | Matching standard errors |
| `fold_0 … fold_4` | string | MPRA folds |

The other 21 columns are original source fields kept under their original names.

## Usage

Example with [DNALLM](https://github.com/zhangtaolab/DNALLM): subset `MPRA_fold_0`, regression, `num_labels=1`.

```python
from modelscope.msdatasets import MsDataset
from dnallm import DNADataset

ds = MsDataset.load("User040218/ReporterNet", subset_name="MPRA_fold_0")
ds = ds.rename_column("HepG2_log2FC_float", "labels")
# tokenizer: the tokenizer of your DNALLM model; set max_length to match its tokenization
data = DNADataset(ds, tokenizer=tokenizer, max_length=602)
```

In MPRA subsets any `*_log2FC_float` column can serve as the label; STARR subsets use `labels`; lentiMPRA subsets use sequence column `seq` and label `mean_value_float`.

## Notes

- The original MPRA log2FC values are stored as text; use the matching `*_float` numeric columns.
- STARR and lentiMPRA subsets have a train split only.
- The data of the three lentiMPRA models are also available as `zhangtaolab/human_lentiMPRA`.

## Source

If you use this dataset, please cite:

- Gosai SJ, Castro RI, Fuentes N, et al. Machine-guided design of cell-type-targeting cis-regulatory elements. Nature 2024;634(8036):1211-1220. doi:10.1038/s41586-024-08070-z
- Agarwal V, Inoue F, Schubach M, et al. Massively parallel characterization of transcriptional regulatory elements. Nature 2025;639(8054):411-420. doi:10.1038/s41586-024-08430-9

## License

Declared as `other`: use of the data follows the terms of the original data providers listed in Source.