BEACON
Collection
Official BEACON/RNABenchmark downstream datasets for RNA sequence, structure, and function tasks. • 13 items • Updated
id stringlengths 6 8 | sequence stringlengths 33 941 | target_file stringlengths 22 24 | target listlengths 33 941 | target_range stringclasses 1
value |
|---|---|---|---|---|
6hiw_CA | "UAAAUUAUGGUCAAUUGUUAGUAUUCAUAUUAAUUUUUUUAAAUGUUUUAUCAUUUUAUAAAGGUUUAUUUUUGAAAGAUUUUUUGUAUAAAAUUUUAG(...TRUNCATED) | contact_map/6hiw_CA.npy | [[true,true,true,false,false,false,false,false,false,false,false,false,false,false,false,false,false(...TRUNCATED) | binary |
5y7m_D | GUAUGGUUAACCCGCAGACGAUCGAAAGGGAUGUGGAUGAAAGCCUAAGCC | contact_map/5y7m_D.npy | [[true,true,true,false,false,false,false,false,false,false,false,false,false,false,false,false,false(...TRUNCATED) | binary |
6eri_Ax | "UUCUGGUGUCCUAGGCGUAGAGGAACCACACCAAUCCAUCCCGAACUUGGUGGUUAAACUCUACUGCGGUGACGAUACUGUAGGGGAGGUCCUGCGGAA(...TRUNCATED) | contact_map/6eri_Ax.npy | [[true,true,true,false,false,false,false,false,false,false,false,false,false,false,false,false,false(...TRUNCATED) | binary |
1i6u_C | GGGCCCGGUAAGUCUCUUCGGAGAUACUGCCGGGCCC | contact_map/1i6u_C.npy | [[true,true,true,false,false,false,false,false,false,false,false,false,false,false,false,false,false(...TRUNCATED) | binary |
4x4v_B | GGCGCUGCGGGGUUCGAGUCCCCGCAGGCCACC | contact_map/4x4v_B.npy | [[true,true,true,false,false,false,false,false,false,false,false,false,false,false,false,false,false(...TRUNCATED) | binary |
3zgz_B | GGCCCGGAUGGUGGAAUCGGUAGACACAAGGGAUUCCCUCGGCGUUCGCGCUGUGCGGGUUCAAGUCCCGCUCCGGGCACCA | contact_map/3zgz_B.npy | [[true,true,true,false,false,false,false,false,false,false,false,false,false,false,false,false,false(...TRUNCATED) | binary |
4v9k_AW | GGCUACGUAGCUCAGUUGGUUAGAGCACAUCACUCAUAAUGAUGGGGUCACAGGUUCGAAUCCCGUCGUAGCCACCA | contact_map/4v9k_AW.npy | [[true,true,true,false,false,false,false,false,false,false,false,false,false,false,false,false,false(...TRUNCATED) | binary |
4xnr_X | GGGAAGAUAUAAUCCUAAUGAUAUGGUUUGGGAGUUUCUACCAAGAGCCUUAAACUCUUGAUUAUCUUCCC | contact_map/4xnr_X.npy | [[true,true,true,false,false,false,false,false,false,false,false,false,false,false,false,false,false(...TRUNCATED) | binary |
6qx9_1 | "AUACUUACCUGGCAGGGGAGAUACCAUGAUCACGAAGGUGGUUUUCCCAGGGCGAGGCUUAUCCAUUGCACUCCGGAUGUGCUGACCCCUGCGAUUUCC(...TRUNCATED) | contact_map/6qx9_1.npy | [[true,true,true,false,false,false,false,false,false,false,false,false,false,false,false,false,false(...TRUNCATED) | binary |
4yye_C | GUUAUAUUAGCUUAAUUGGUAGAGCAUUCGUUUUGUAAUCGAAAGGUUUGGGGUUCAAAUCCCUAAUAUAACAC | contact_map/4yye_C.npy | [[true,true,true,false,false,false,false,false,false,false,false,false,false,false,false,false,false(...TRUNCATED) | binary |
End of preview. Expand in Data Studio
BEACON — ContactMap
Nucleotide-level RNA contact-map prediction with binary pairwise targets.
Official data from the shared BEACON/RNABenchmark Drive folder:
https://drive.google.com/drive/folders/19ddrwI8ycvIxkgSV3gDo_VunLofYd4-6?hl=en.
This repository is the standardized Hugging Face publication of the official
task data. The data/ directory is the canonical viewer-friendly layer, and
the original files and directory layout are preserved for reproducibility.
beacon_manifest.json records source file IDs, checksums, sizes, split counts,
schema, and processing notes. Interrupted-upload artifacts ending in
.partial are retained only in the GPUHub raw staging area and excluded from
the published repository.
Canonical splits
| Split | Rows | File |
|---|---|---|
| train | 188 | data/train.parquet |
| validation | 23 | data/validation.parquet |
| test | 80 | data/test.parquet |
| test_RFAM19 | 19 | data/test_RFAM19.parquet |
| test_DIRECT | 6 | data/test_DIRECT.parquet |
Standardized schema
| Field | Arrow type |
|---|---|
id |
string |
sequence |
string |
target_file |
string |
target |
list<element: list<element: bool>> |
target_range |
string |
Standardization notes
- Canonical split names are train, validation, and test; source val.csv/val_1.json is published as validation.
- The official raw directory tree is preserved alongside the standardized data/*.parquet files.
- Source file names and target paths are retained in beacon_manifest.json for provenance.
- Parquet target is a boolean pairwise contact matrix; RFAM19 and DIRECT are retained as separate test splits.
Usage
from datasets import load_dataset
dataset = load_dataset("jiahaozhang2003/beacon-contact-map")
Reproducibility
This repository was generated on GPUHub from the public Drive source with the RiboLoop downstream benchmark data-staging script. No model outputs or training artifacts are included.
- Downloads last month
- 124