Cis-variant Gene-expression Association Analysis: A Targeted Approach to Identify Functional Cis-regulatory Variation

Authors

• Sofiya Arora, University of Nebraska-LincolnFollow

First Advisor

James C. Schnable

Committee Members

Joe Louis, Brandi Sigmon

Date of this Version

5-2026

Document Type

Thesis

Citation

A thesis presented to the faculty of the Graduate College at the University of Nebraska in partial fulfilment of requirements for the degree of Master of Science

Major: Agronomy

Under the supervision of Professor James C. Schnable

Lincoln, Nebraska, May 2026

Comments

Copyright 2026, Sofiya Arora. Used by permission

Abstract

Phenotypic diversity in plants is largely driven by changes in gene function arising from mutations in both the coding and regulatory regions. Regulatory variation is particularly pervasive because it can fine-tune gene expression without disrupting protein function and is therefore less likely to be strongly selected against in natural populations. As a result, it has played a central role in evolution and domestication. In our previous genome-wide eQTL mapping study in maize, nearly two-thirds of the expressed genes were found to harbor detectable cis-expression quantitative trait loci (cis-eQTLs) with strong local effects on gene expression. Though these cis-variants are widespread, it remains unclear which cis-variants actually matter and have a functional impact. To test this, we developed a targeted approach, Cis-variant Gene-expression Association (CGA) analysis, which tests individual cis-eQTLs against all the expressed genes across a population-scale RNA-seq dataset. Unlike traditional eQTL mapping, where each gene is tested against at least 10⁷ genetic markers, CGA reverses the framework by testing each variant of interest against the expression of approximately 10⁴ genes, thereby reducing the multiple testing burden and increasing the power to detect subtle downstream effects. Validation of the approach using 13 cis-eQTL markers with previously identified trans-effects revealed that CGA recapitulates known associations while identifying approximately 10-fold more trans-signals. Extending this analysis to 331 cis-eQTLs associated with classical maize mutants showed that, although cis-regulatory variation is widespread, only a small subset exhibits substantial downstream influence. Specifically, only ~4% of tested cis-eQTLs showed a significantly elevated number of trans-associations compared to a null distribution derived from matched random markers. Functional enrichment analysis of top loci, including RAD17 and RGD2, demonstrated that downstream gene sets are biologically coherent and consistent with known functions of focal genes, DNA repair and RNA silencing pathways, respectively. Together, these results suggest that most cis-regulatory variation is likely neutral or buffered within gene regulatory networks, and CGA provides a scalable framework to identify functionally impactful regulatory variants by quantifying their downstream transcriptional effects.

Advisor: James C. Schnable