PlantHelixSeek-Anno

# PlantHelixSeek-Anno Dataset Fine-tuning dataset for **gene structure prediction** — 17-class BILOU token classification. Used to train [PlantHelixSeek-Anno](https://huggingface.co/zhangtaolab/PlantHelixSeek-Anno). ## Format Parquet, 7 columns per row: | Column | Type | Description | |--------|------|-------------| | `species` | string | Species name | | `chrom` | string | Chromosome | | `start` | int64 | Window start coordinate | | `end` | int64 | Window end coordinate | | `strand` | string | Strand (`+` / `-`) | | `sequence` | string | DNA sequence (8192 bp window) | | `labels` | list\<int32\> | Per-base 17-class BILOU labels (same length as sequence) | ## Splits | Split | Rows | Size | |-------|-----:|-----:| | train | 875,050 | 3.1 GB | | val | 133,061 | 475 MB | ## Label scheme (17-class BILOU) `O` (intergenic), `B/I/L/U-CDS`, `B/I/L/U-INTRON`, `B/I/L/U-UTR5`, `B/I/L/U-UTR3` ## License CC-BY-NC-4.0