ENCODE Regulatory Elements Identify Rare Variant Associations Across Complex Traits

In a whole-genome sequencing study published in BMC Genomics, researchers demonstrated that candidate cis-regulatory elements can serve as an effective framework for identifying trait-associated regulatory regions driven by rare genetic variants. Led by authors Jack Thomas Flanagan, Kisung Nam, and Seunggeun Lee, the research team analyzed twenty complex traits to investigate the role of rare non-coding variants in trait heritability. Using candidate cis-regulatory elements from the ENCODE project as units of analysis, the researchers conducted rare-variant association testing in an initial discovery cohort of 150,000 White-British participants. In this discovery cohort, the analysis successfully identified 136 genomic loci that captured 699 statistically significant associations, showing that rare variants in regulatory regions contribute to complex trait variation.

The authors noted that large-scale whole-genome sequencing cohorts provide an unprecedented opportunity to explore the role of rare variants in non-coding regions in complex traits. However, unlike gene-based exome analysis, non-coding regions have consistently lacked a well-defined unit for rare-variant testing. Because non-coding regions capture the majority of reported genome-wide association study loci, establishing systematic approaches to efficiently investigate the contribution of rare variants within these regions to trait heritability represents an important scientific need. Without clearly defined testing units, evaluating aggregate rare-variant effects across the expansive non-coding genome has remained a significant methodological hurdle in human genomics.

To establish a functional unit for testing, the researchers utilized 1,036,913 candidate cis-regulatory elements, abbreviated as cCREs, from the ENCODE encyclopaedia as analysis units. With these defined units in place, the investigators conducted rare-variant association analysis across twenty distinct traits using the SAIGE-GENE+ method. In the discovery dataset comprising 150,000 White-British samples, this cCRE-based framework identified 136 loci capturing 699 significant associations. The authors observed that aggregating rare variants within candidate cis-regulatory elements enabled the detection of non-coding regulatory signals across multiple traits that would otherwise lack a clear analytical grouping.

The association signals identified in the discovery phase demonstrated strong reproducibility when tested in an independent sample cohort. Evaluating the discovery findings within a validation set of 250,000 White-British samples, the investigators replicated 96 percent of the association signals. In addition to high replication across cohorts, the observed associations exhibited distinct functional enrichment patterns. Specifically, the trait associations were enriched in promoter-like signals, referred to as PLS, as well as proximal enhancer-like signals, referred to as pELS, implicating both promoter and enhancer elements in trait heritability.

To determine whether the identified associations represented independent regulatory effects, the research team carried out detailed conditional analyses across the 136 loci across all twenty traits. The investigators conditioned the loci on known gene associations in addition to genome-wide association study variants, and subsequently performed intra-cCRE conditional analysis. Following these conditional testing steps, the authors identified 33 loci capturing 54 independent cCRE-driven association signals. To demonstrate the practical effectiveness of cCRE-based analysis, the authors presented a leukemia case study that highlighted key loci, including a potential promoter-like signal for SRSF2 and a proximal enhancer-like signal for BCL6.

In their conclusion, the authors stated that candidate cis-regulatory elements provide an efficient framework for testing the aggregate effects of rare variants within non-coding regions in whole-genome sequencing cohorts. The analysis identified multiple regions containing potentially novel association loci, implicating both enhancer and promoter elements in trait heritability. Crucially, using cCREs as analysis units supports the integration of functional annotations from the ENCODE encyclopaedia, facilitating the linking of association loci to their putative target genes based on evidence of regulatory activity. This approach enhances the biological interpretability of association signals and supports further downstream investigation of enhancer-gene interactions. The study was conducted in discovery and validation cohorts of White-British samples.

Subject of Research: Rare-variant association analysis of non-coding regulatory regions in whole-genome sequencing

Article Title: ENCODE guided WGS analysis can identify trait associated regulatory regions driven by rare variants

Article References: Flanagan, J. T., Nam, K., & Lee, S. (2026). ENCODE guided WGS analysis can identify trait associated regulatory regions driven by rare variants. BMC Genomics. https://doi.org/10.1186/s12864-026-13479-z

Image Credits: AI Generated

DOI: 10.1186/s12864-026-13479-z

Keywords: whole-genome sequencing, candidate cis-regulatory elements, rare variants, complex traits, ENCODE, genomics, SAIGE-GENE+, non-coding genome, guided, analysis, identify, trait