A linear-transformer hybrid for single-nucleotide polymorphism-based genotype-to-phenotype prediction in grapevine.
- 1Department of Biological and Agricultural Engineering, TX A&M AgriLife Research, Texas A&M University System, Dallas, Texas, USA.
- 2Department of Soil and Crop Sciences, TX A&M AgriLife Research, Texas A&M University System, Dallas, Texas, USA.
- 3USDA-ARS, Grape Genetics Research Unit, Geneva, New York, USA.
Abstract
Robust genotype-to-phenotype (G2P) prediction is essential for accelerating grapevine (Vitis spp.) breeding decisions and genetic gain. However, it remains challenging to measure complex traits under variable field conditions and across years. In this study, we propose a linear-Transformer (LiT)-G2P, an automated predictive framework that integrates additive genetic variance effects with Transformer-based nonlinear interactions using genome-wide single-nucleotide polymorphisms (SNPs) data. We evaluated LiT-G2P on a panel of diverse grape accessions, genotyped with SNP markers and measured for phenotypes across three consecutive years. Target phenotypic traits include leaf hair density and trichome density of grapevines. Across both single-year and cross-year testing scenarios, LiT-G2P consistently improves prediction performance compared with baseline models. For leaf hair density, LiT-G2P achieved the lowest prediction errors in both single-year and cross-year evaluations, with root mean square error (RMSE) values of 0.475 and 0.460 and ± 0.5 accuracies of 78.6% and 72.9%, respectively. For trichome density, LiT-G2P achieved RMSE values of 0.637 and 0.766 and corresponding accuracies of 62.6% and 58.2%. In addition, we extracted model-prioritized SNPs from attention weights and applied genotype-stratified analysis to provide interpretable candidate marker for downstream validation. These results support practical SNP-based predictive modeling for genomic selection and identify the remaining need for more detailed field data, larger multi-year grapevine panels, and independent grapevine validation before broader deployment.
Related
- VitisGP: Multi-Task Benchmark for Genotype to Phenotype Grapevine Prediction
- Canonical correlation analysis identifies single nucleotide polymorphism combinations and gene expression associated with membranous nephropathy
- CCR5Δ32 Polymorphism and Inflammatory Bowel Disease: A Case–Control and Genotype–Phenotype Study in a Polish Population
- DeepCDS: Ab initio coding sequence prediction in prokaryotic short nucleotide sequences
- LKMHCpan: Transformer-optimized HLA-I pseudo-sequences enhance neoantigen prediction for immunotherapy
BibTeX
@article{wang2026lineartransformer,
title = {{A linear-transformer hybrid for single-nucleotide polymorphism-based genotype-to-phenotype prediction in grapevine.}},
author = {Yibin Wang and Murukarthick Jayakodi and Silvas Kirubakaran and Ambika Chandra and Azlan Zahid},
journal = {The plant genome},
year = {2026},
month = nov,
doi = {10.1002/tpg2.70335},
url = {https://europepmc.org/article/MED/42972217},
note = {CC BY},
}
RIS
TY - JOUR
TI - A linear-transformer hybrid for single-nucleotide polymorphism-based genotype-to-phenotype prediction in grapevine.
AU - Yibin Wang
AU - Murukarthick Jayakodi
AU - Silvas Kirubakaran
AU - Ambika Chandra
AU - Azlan Zahid
PY - 2026
DA - 2026/11/02
JO - The plant genome
DO - 10.1002/tpg2.70335
UR - https://europepmc.org/article/MED/42972217
AB - Robust genotype-to-phenotype (G2P) prediction is essential for accelerating grapevine (Vitis spp.) breeding decisions and genetic gain. However, it remains challenging to measure complex traits under variable field conditions and across years. In this study, we propose a linear-Transformer (LiT)-G2P, an automated predictive framework that integrates additive genetic variance effects with Transformer-based nonlinear interactions using genome-wide single-nucleotide polymorphisms (SNPs) data. We evaluated LiT-G2P on a panel of diverse grape accessions, genotyped with SNP markers and measured for phenotypes across three consecutive years. Target phenotypic traits include leaf hair density and trichome density of grapevines. Across both single-year and cross-year testing scenarios, LiT-G2P consistently improves prediction performance compared with baseline models. For leaf hair density, LiT-G2P achieved the lowest prediction errors in both single-year and cross-year evaluations, with root mean square error (RMSE) values of 0.475 and 0.460 and ± 0.5 accuracies of 78.6% and 72.9%, respectively. For trichome density, LiT-G2P achieved RMSE values of 0.637 and 0.766 and corresponding accuracies of 62.6% and 58.2%. In addition, we extracted model-prioritized SNPs from attention weights and applied genotype-stratified analysis to provide interpretable candidate marker for downstream validation. These results support practical SNP-based predictive modeling for genomic selection and identify the remaining need for more detailed field data, larger multi-year grapevine panels, and independent grapevine validation before broader deployment.
SN - 1940-3372
ER -