Adversarial random forests for omics synthesis
- 1Emory University
Abstract
Data availability is critical for understanding complex disease pathways and developing robust predictive models. Although high-throughput omics technologies have improved insight into disease mechanisms, data acquisition from inaccessible tissues such as the central nervous system remains a major limitation, causing small sample sizes and complicating early prediction of neurodegenerative disorders such as Alzheimer's and Parkinson's diseases. Generative modeling has emerged as a powerful approach for synthesizing data to support downstream clustering and prediction with small sample size, but existing methods rarely handle high-dimensional tabular omics data effectively. Adversarial random forests (ARFs) provide a well-performing framework for tabular data generation but are not designed for high-dimensional settings. To address this limitation, we introduce high-dimensional ARF (h-ARF), an extension of ARF optimized for integrated clinical and high-dimensional omics data. Using benchmarks across nine datasets and eight performance metrics, we show that h-ARF better preserves both feature distributions, and downstream clustering and prediction utilities compared with ARFs. The method is implemented in the open-source R package harf, available on CRAN.
Related
- Generative Adversarial Networks for Omics Data: An Overview and Future Directions for Livestock Applications
- Ecophysiological variation of transpiration of pine forests: synthesis of new and published results
- Multi-granularity Adversarial Generation Integrated Consistency Representation for Chest Low-Contrast-Enhanced CT Synthesis
- Omics data discovery agents: Agent-supported retrieval, reanalysis, and synthesis of published omics data
- NeuWalk: A Python Library for Data-Driven Synthesis of Neuron Morphologies Using Random Walks
BibTeX
@misc{fouodo2026adversarial,
title = {{Adversarial random forests for omics synthesis}},
author = {C. J. Kuete Fouodo and J. Kapar and A. Huels and D. Liang and M. N. Wright},
howpublished = {bioRxiv},
year = {2026},
month = sep,
doi = {10.64898/2026.09.09.750490},
url = {https://www.biorxiv.org/content/10.64898/2026.09.09.750490v1},
note = {CC BY-NC-ND},
}
RIS
TY - UNPB
TI - Adversarial random forests for omics synthesis
AU - C. J. Kuete Fouodo
AU - J. Kapar
AU - A. Huels
AU - D. Liang
AU - M. N. Wright
PY - 2026
DA - 2026/09/09
PB - bioRxiv
DO - 10.64898/2026.09.09.750490
UR - https://www.biorxiv.org/content/10.64898/2026.09.09.750490v1
AB - Data availability is critical for understanding complex disease pathways and developing robust predictive models. Although high-throughput omics technologies have improved insight into disease mechanisms, data acquisition from inaccessible tissues such as the central nervous system remains a major limitation, causing small sample sizes and complicating early prediction of neurodegenerative disorders such as Alzheimer's and Parkinson's diseases. Generative modeling has emerged as a powerful approach for synthesizing data to support downstream clustering and prediction with small sample size, but existing methods rarely handle high-dimensional tabular omics data effectively. Adversarial random forests (ARFs) provide a well-performing framework for tabular data generation but are not designed for high-dimensional settings. To address this limitation, we introduce high-dimensional ARF (h-ARF), an extension of ARF optimized for integrated clinical and high-dimensional omics data. Using benchmarks across nine datasets and eight performance metrics, we show that h-ARF better preserves both feature distributions, and downstream clustering and prediction utilities compared with ARFs. The method is implemented in the open-source R package harf, available on CRAN.
ER -