Gene function & annotation

Functional Validation and Predictive Modeling of Chicken Enhancers and Regulatory Variants Using Massively Parallel Reporter Assays

---

Abstract

The Functional Annotation of Animal Genomes (FAANG) consortium has substantially expanded the catalog of putative enhancers in the chicken genome, yet the functional activity of these elements and the specific causal variants they harbor remain largely unverified. To address this gap, we established a Massively Parallel Reporter Assay (MPRA) in chicken DF-1 cells to systematically assess enhancer activity and allele-specific regulatory effects. We initially screened 999 oligonucleotides representing 680 allele-specific expression (ASE)-associated sequences, together with 190 positive and 100 negative controls. The assay demonstrated excellent technical performance, achieving a 99.9% capture rate and high reproducibility, with 95.9% of barcodes correctly matched to their input oligos. Of the 340 variants tested, 26% showed regulatory activity, and 82% of these functioned as enhancers. Seven variants exhibited significant allelic imbalance, including rs14718121, where the C allele increased reporter expression by 1.32-fold (FDR = 3.9 à— 10⁻⁵). Predicted target genes of this variant were enriched for cell-differentiation and immune-signaling pathways, highlighting potential roles in developmental and immune-related processes. To complement experimental validation, we trained gapped k-mer support vector machine (gkm-SVM) models using chromatin features (H3K4me3, H3K27ac, ATAC-seq, H3K4me1, H3K27me3) from DF-1 cells. These models demonstrated robust predictive performance (AUC up to 0.88 for H3K4me3) and substantial concordance with MPRA results (maximum Spearman's ρ = 0.55), demonstrating that enhancer activity can be inferred from DNA sequence features alone. Together, these findings establish an integrated framework combining high-throughput experimental validation and machine-learning prediction to prioritize functional regulatory elements in the chicken genome. This approach enables systematic identification of causal variants underlying complex traits and enhances the biological interpretation of GWAS signals, with direct implications for genomic selection in poultry. We are currently scaling this MPRA platform to approximately 15,000 sequences, including FAANG-predicted enhancers across multiple tissues and regulatory variants associated with allele-specific expression during Marek's disease virus infection. This expanded effort will generate a comprehensive functional atlas of chicken enhancers and regulatory variants, advancing our understanding of regulatory genomics and complex trait architecture in poultry.

Keywords: 2026

How to Cite:

An, L., Wang, Y. & Zhou, H., (2026) “Functional Validation and Predictive Modeling of Chicken Enhancers and Regulatory Variants Using Massively Parallel Reporter Assays”, World Congress on Genetics Applied to Livestock Production Digital Archive 2026(1): 2295799. doi: https://doi.org/10.31274/wcgalp.24371

Rights: 1