Benchmarking ensemble and deep learning models for drought-responsive gene prioritisation in rice using multi-omics data
4 affiliations
- 1Institute of Biological Sciences, Faculty of Science, Universiti Malaya, Kuala Lumpur, Wilayah Persekutuan Kuala Lumpur, Malaysia
- 2Institute of Mathematical Sciences, Faculty of Science, Universiti Malaya, Kuala Lumpur, Wilayah Persekutuan Kuala Lumpur, Malaysia
- 3Biotechnology and Nanotechnology Research Centre, Malaysian Agricultural Research and Development Institute (MARDI), Serdang, Selangor, Malaysia
- 4Centre of Research in Systems Biology, Structural Bioinformatics and Human Digital Imaging, Universiti Malaya, Kuala Lumpur, Malaysia
Abstract
Identifying drought-responsive genes is critical for climate-resilient crop breeding, yet computational gene–trait association often relies on single-omics or generalized genomic datasets. Optimal algorithm selection for high-dimensional, small multi-omics datasets also remains a persistent challenge. We present a machine learning framework that extends the QTG-Finder feature set by integrating transcriptomic and proteomic features to classify drought-responsive genes in rice. Using MBKBase annotations, we evaluated the ability of these models to prioritise 336 drought-responsive genes among 2,416 abiotic stress-associated genes, of which the remaining 2,080 constituted an operational background set. We tested four algorithms: Random Forest, XGBoost, one-dimensional convolutional neural network, and multilayer perceptron, while addressing severe class imbalance and assessing model stability. Tree-based ensemble methods achieved higher performance than the deep learning models in this dataset, which showed early overfitting. The fine-tuned Random Forest achieved a recall of 0.67 and an AUC-ROC of 0.87 under nested cross-validation, conditional on a feature matrix assembled before gene-level data splitting. Across repeated evaluations of a small internal holdout of 10 genes, mean accuracy and recall were 0.90 and 0.63; this set is too small to support a generalizability claim and is reported only as a consistency check. The newly added transcriptomic and proteomic features improved recall from 0.47 to 0.67 compared with genomics alone. Feature importance and ablation analyses indicated that transcriptomics-derived features, particularly measures of expression distribution, magnitude, and direction, contributed most to predictive performance. Functional enrichment analysis supported the biological relevance of the prioritised genes. These results demonstrate the value of multi-omics integration for gene prioritisation in small tabular datasets while highlighting current limitations of deep learning in this setting.
Related
- HBOCpred: An Ensemble Framework for Prioritisation of Germline Missense Variants in Hereditary Breast and Ovarian Cancer
- Experimental benchmarking of in silico annotations at prostate cancer risk loci and hypothesis-generating cross-cancer prioritisation of ovarian risk variants.
- CYSTM1, a drought-responsive cysteine-rich transmembrane module protein, negatively affects drought tolerance in Arabidopsis thaliana
- A phosphorylation-gated GAPDH-MAPK6-LBD1 module controls drought-responsive transcription
- Genome-wide characterization and drought-responsive expression profiling of plasma membrane intrinsic proteins (PIPs) in Caragana korshinskii
BibTeX
@article{govaichelvan2026benchmarking,
title = {{Benchmarking ensemble and deep learning models for drought-responsive gene prioritisation in rice using multi-omics data}},
author = {Kumanan N. Govaichelvan and Dharini Pathmanathan and Rabiatul Adawiah Zainal-Abidin and Arpah Abu},
journal = {PLOS One},
year = {2026},
month = oct,
doi = {10.1371/journal.pone.0360052},
url = {https://doi.org/10.1371/journal.pone.0360052},
note = {CC BY 4.0},
}
RIS
TY - JOUR
TI - Benchmarking ensemble and deep learning models for drought-responsive gene prioritisation in rice using multi-omics data
AU - Kumanan N. Govaichelvan
AU - Dharini Pathmanathan
AU - Rabiatul Adawiah Zainal-Abidin
AU - Arpah Abu
PY - 2026
DA - 2026/10/09
JO - PLOS One
DO - 10.1371/journal.pone.0360052
UR - https://doi.org/10.1371/journal.pone.0360052
AB - Identifying drought-responsive genes is critical for climate-resilient crop breeding, yet computational gene–trait association often relies on single-omics or generalized genomic datasets. Optimal algorithm selection for high-dimensional, small multi-omics datasets also remains a persistent challenge. We present a machine learning framework that extends the QTG-Finder feature set by integrating transcriptomic and proteomic features to classify drought-responsive genes in rice. Using MBKBase annotations, we evaluated the ability of these models to prioritise 336 drought-responsive genes among 2,416 abiotic stress-associated genes, of which the remaining 2,080 constituted an operational background set. We tested four algorithms: Random Forest, XGBoost, one-dimensional convolutional neural network, and multilayer perceptron, while addressing severe class imbalance and assessing model stability. Tree-based ensemble methods achieved higher performance than the deep learning models in this dataset, which showed early overfitting. The fine-tuned Random Forest achieved a recall of 0.67 and an AUC-ROC of 0.87 under nested cross-validation, conditional on a feature matrix assembled before gene-level data splitting. Across repeated evaluations of a small internal holdout of 10 genes, mean accuracy and recall were 0.90 and 0.63; this set is too small to support a generalizability claim and is reported only as a consistency check. The newly added transcriptomic and proteomic features improved recall from 0.47 to 0.67 compared with genomics alone. Feature importance and ablation analyses indicated that transcriptomics-derived features, particularly measures of expression distribution, magnitude, and direction, contributed most to predictive performance. Functional enrichment analysis supported the biological relevance of the prioritised genes. These results demonstrate the value of multi-omics integration for gene prioritisation in small tabular datasets while highlighting current limitations of deep learning in this setting.
SN - 1932-6203
ER -