PlantGeneAnn: a high-accuracy ab initio plant gene annotation tool fine-tuned from a genome foundation model
- 1Huazhong Agricultural University
Abstract
Structural annotation remains a bottleneck for translating the growing catalog of high-quality plant genomes into biological insights, as current pipelines rely on RNA-seq and homologous protein evidence that are often unavailable for non-model species. Existing ab initio tools are constrained by model scale and perform poorly in complex plant genomic regions. We present PlantGeneAnn, a plant-optimized ab initio gene annotation tool fine-tuned from a genome foundation model. It combines a Mamba-MoE DNA encoder with a U-Net-style embedding decoder and a semi-Markov decoding pipeline. Trained on 40 angiosperm species, PlantGeneAnn establishes a new state-of-the-art (SOTA) across a stringent 20-species benchmark, significantly outperforming Helixer, Tiberius, and ANNEVO at exon-, intron-, and locus-levels, with mean F1 scores of 0.856, 0.880, and 0.707, respectively. It also enables efficient customized fine-tuning: LoRA fine-tuning increased the locus-level F1 of the out-of-distribution bryophyte Physcomitrium patens from 0.581 to 0.736, with performance plateauing at 40% of the fine-tuning data. Notably, PlantGeneAnn requires substantially lower peak memory usage than other baseline tools, allowing annotation of large plant genomes on standard personal workstations. PlantGeneAnn thus provides accurate, hardware-friendly ab initio annotation for diverse plant genomes, particularly newly assembled genomes, population-scale projects, and non-model plants.
Related
- Where single-cell foundation models help: a benchmark of annotation, perturbation and regulatory inference against simple baselines
- Thirteen quick tips for bacteriophage genome annotation
- Cell-type separability predicts annotation accuracy and outweighs algorithm choice: a factorial benchmark across seven scRNA paradigms
- Genome annotation of a cluster FD Arthrobacter bacteriophage Casab isolated from soil in Tennessee.
- A high-quality chromosome-level genome assembly and annotation of the medicinal orchid Cremastra appendiculata (Orchidaceae)
BibTeX
@misc{qizhe2026plantgeneann,
title = {{PlantGeneAnn: a high-accuracy ab initio plant gene annotation tool fine-tuned from a genome foundation model}},
author = {Z. Qizhe and Z. Zhengyang and L. Kepeng and J. Wang and D. Kaixuan and X. Wei and Y. Qing-Yong and H. Xuehai},
howpublished = {bioRxiv},
year = {2026},
month = jun,
doi = {10.64898/2026.06.25.733695},
url = {https://www.biorxiv.org/content/10.64898/2026.06.25.733695v2},
note = {CC BY-NC-ND},
}
RIS
TY - UNPB
TI - PlantGeneAnn: a high-accuracy ab initio plant gene annotation tool fine-tuned from a genome foundation model
AU - Z. Qizhe
AU - Z. Zhengyang
AU - L. Kepeng
AU - J. Wang
AU - D. Kaixuan
AU - X. Wei
AU - Y. Qing-Yong
AU - H. Xuehai
PY - 2026
DA - 2026/06/25
PB - bioRxiv
DO - 10.64898/2026.06.25.733695
UR - https://www.biorxiv.org/content/10.64898/2026.06.25.733695v2
AB - Structural annotation remains a bottleneck for translating the growing catalog of high-quality plant genomes into biological insights, as current pipelines rely on RNA-seq and homologous protein evidence that are often unavailable for non-model species. Existing ab initio tools are constrained by model scale and perform poorly in complex plant genomic regions. We present PlantGeneAnn, a plant-optimized ab initio gene annotation tool fine-tuned from a genome foundation model. It combines a Mamba-MoE DNA encoder with a U-Net-style embedding decoder and a semi-Markov decoding pipeline. Trained on 40 angiosperm species, PlantGeneAnn establishes a new state-of-the-art (SOTA) across a stringent 20-species benchmark, significantly outperforming Helixer, Tiberius, and ANNEVO at exon-, intron-, and locus-levels, with mean F1 scores of 0.856, 0.880, and 0.707, respectively. It also enables efficient customized fine-tuning: LoRA fine-tuning increased the locus-level F1 of the out-of-distribution bryophyte Physcomitrium patens from 0.581 to 0.736, with performance plateauing at 40% of the fine-tuning data. Notably, PlantGeneAnn requires substantially lower peak memory usage than other baseline tools, allowing annotation of large plant genomes on standard personal workstations. PlantGeneAnn thus provides accurate, hardware-friendly ab initio annotation for diverse plant genomes, particularly newly assembled genomes, population-scale projects, and non-model plants.
ER -