Fine-tuned from PlantHelixSeek for gene structure prediction — labels every nucleotide with one of 17 BILOU tags spanning CDS / intron / 5'UTR / 3'UTR / intergenic.

Task

Type: Token classification (17-class BILOU)
Base model: PlantHelixSeek (470M)

Label scheme

Group	Tags
intergenic	`O`
CDS	`B-CDS` `I-CDS` `L-CDS` `U-CDS`
Intron	`B-INTRON` `I-INTRON` `L-INTRON` `U-INTRON`
5'UTR	`B-UTR5` `I-UTR5` `L-UTR5` `U-UTR5`
3'UTR	`B-UTR3` `I-UTR3` `L-UTR3` `U-UTR3`

Usage

python

from transformers import AutoModelForTokenClassification, AutoTokenizer

model = AutoModelForTokenClassification.from_pretrained(
    "zhangtaolab/PlantHelixSeek-Anno", trust_remote_code=True
)
tokenizer = AutoTokenizer.from_pretrained(
    "zhangtaolab/PlantHelixSeek-Anno", trust_remote_code=True
)

Full genome-wide prediction + GFF3 decoding pipeline: see scripts/gene_annotation/.

License

CC-BY-NC 4.0