An introduction to the Illumina 5-base solution

Introduction

DNA molecules contain multiple layers of information, including genetic sequences and epigenetic modifications, making them inherently multiomic. In addition to the four canonical nucleotides, there is a fifth base: methylated cytosine (Figure 1). The methylated cytosines are epigenetic tags that act as dynamic molecular signals to modify gene expression in response to environmental, developmental, and cellular factors. Much like genetic variants provide insight into health states, aberrant methylation signatures can serve as indicators of disease. Layering genetic and epigenetic information together can enhance resolution and sensitivity relative to single-omic studies alone, revealing new biological insights.

Commercially available methylation profiling methods such as bisulfite or EM-seq convert unmethylated cytosines to thymine, removing the majority of cytosines from the library and resulting in a low-complexity genome that is challenging to sequence and align. Methylated cytosines are then inferred from the remaining unconverted cytosines. While these technologies have been fundamental tools in shaping our current understanding of DNA methylation, they lack sufficient accuracy to also enable high-resolution genetic variant calling. We envisioned that a solution offering detection of both variant types from the same workflow would greatly accelerate discoveries.

In that context, we introduce an innovative 5-base solution that enables methylation profiling and genetic variant calling in a fast, simple, and comprehensive workflow relative to alternative approaches (Figure 2).

A novel solution

At the core of the Illumina 5-base solution is a novel, bespoke enzyme that was engineered in-house at Illumina for selective and direct conversion of methylated cytosines to sequencer-ready thymine (Figure 3). Millions of protein variants were screened, over iterative generations, to identify a protein with the highest selectivity for methylated cytosine and efficiency such that the reaction performs optimally in only 30 minutes.

Foundational to the 5-base solution are the highly optimized Illumina DRAGEN algorithms that perform methylation-aware genome alignment, methylation quantification, and high-accuracy variant calling in a single analysis. A fundamental innovation of this software is the ability to distinguish a methyl-converted cytosine from a  C>T single-nucleotide variant (SNV) by leveraging the information encoded in the complementary DNA strand. For a methylated cytosine, the pipeline expects the complimentary base to be a guanine, whereas for a  C>T variant, the complimentary base is adenosine (Figure 4). The methylation level and variant allele frequency at an individual cytosine position are inferred from the information contained in reads covering both strands.

How it works

The methyl cytosine-conversion reaction is a simple, integrated step within a high-efficiency ligation-based library preparation workflow. The workflow begins with extracted, sheared genomic DNA (gDNA) or cell-free DNA (cfDNA), which then undergoes end-repair, A-tailing, and adapter ligation. Next, the DNA is denatured (20 minutes) and incubated with the 5-methyl cytosine conversion enzyme (30 minutes) for direct conversion of methylated cytosines in a single reaction. After conversion, samples are amplified by PCR and bead purified for sequencing on standard lllumina sequencing platforms. The library preparation is highly streamlined and can be completed in less than eight hours. The conversion chemistry is also compatible with targeted capture enrichment for high-depth applications, including UMI-based error correction.

Once sequencing is complete, the alignment, methylation annotation, and variant calling can be performed with a single push-button analysis pipeline in as little as an hour. The results can be further processed with interpretation software, such as Illumina Connected Multiomics and Emedgene.

High-accuracy methylation profiling with less sequencing

In benchmarking studies, the 5-base solution provides methylation profiling with comparable accuracy to gold standard methylation conversion methods. The chemistry achieves high conversion rates of methylated cytosines and very low off-target conversion of unmethylated cytosines based on evaluation with methylated and unmethylated pUC and lambda control genomes (Figure 6A). Methylation levels measured with the 5-base solution across CpG islands in human genomes are highly consistent with values obtained from EM-seq conversion. The high mappability of 5-base libraries from converting only the methylated cytosines results in more CpG sites covered per sequencing depth (Figure 6C), which translates to a higher discovery power with the 5-base solution. 

Simultaneous detection of genetic variants with high sensitivity

Methylation chemistries that convert unmethylated cytosine to thymine pose challenges for genetic variant detection and typically require that standard sequencing libraries be processed separately if genetic variant information is desired. The Illumina 5-base solution eliminates the need for two separate library preparations by enabling high-resolution variant detection at all four base positions (A, T, G, and C) in addition to methylated cytosine detection within the same sequencing read. When analyzed with DRAGEN applications available in Illumina BaseSpace Sequencing Hub or Illumina Connected Analytics, the 5-base solution delivers small variant calling comparable to the accuracy of standard whole-genome sequencing—for both germline and somatic variants (Figures 7A, 7B). Additionally, the technology can be combined with targeted capture enrichment for detection of low-frequency variants from blood, fresh-frozen tissue, or cfDNA (Figure 7C).

One assay with dual-omic insights empowers discovery

The Illumina 5-base solution is a novel approach for harnessing genetic and epigenetic insights from a single workflow that we anticipate will enhance discovery power across broad research applications, including cancer and rare disease. The simultaneous reporting of genome and methylome status on a single read can be leveraged to link pathogenic mutations to somatic allele-specific methylation (Figure 8A). We also demonstrate that the 5-base solution can resolve rare disease cases that cannot be resolved by standard WGS alone and require dual-omic interrogation with methylation signatures (Figure 8B). At Illumina, we are continuing to explore how combining dual-omic insights will drive new discoveries across a wide range of exciting applications. We anticipate that the 5-base solution will be available commercially in early 2026.