We are interested in determining how phenotypic variation is driven by natural and spontaneous epigenetic variation. Recent advances in genomic technologies are enabling acquisition of sequence-level data at an unprecedented rate and resolution. As a result, genomes for thousands of individuals are being analyzed to determine the total genetic variation present within a species and to determine the impact these variants have on phenotypic variation. Missing from these efforts is the identification of environmentally induced, spontaneous and natural epigenetic alleles (epialleles).

We apply epigenomic approaches to populations to study the impact that epialleles have on life history traits and to understand their role in establishing responses to the environment. These epialleles are often associated with changes in their DNA methylation status making their identification possible. Systematically identifying these epialleles, understanding their patterns of heritability, their interaction with genetic variants and their specific roles in controlling gene expression is necessary for a comprehensive understanding of phenotypic variation.

Stacks Image 30

Natural variation of DNA methylation in a population of Arabidopsis thaliana.
Above is a representative image from a genome browser displaying whole-genome bisulfite sequencing data.

We are also applying epigenomic profiling methods to discover non-coding regulatory elements across numerous plant species. These methods include genome-wise assays to functionally evaluate identified regions using both gain-of-function and loss-of-function approaches. With advances in genome-editing technologies we are beginning to engineer non-coding regulatory elements as well as to use identified sequences to create modular units that can be used to drive transgene expression in a predictable manner.

Stacks Image 62

Combining nuclei sorting with ATAC-seq reveals regions of chromatin accessibility and candidate cis-regulatory elements in plant genomes. The distance of accessible chromatin and their associated cis-regulatory elements to their nearest genes is greater in plants with larger genomes.


Stacks Image 43
The term "epigenetic" was originally defined to convey how an organism of a single genotype can develop from undifferentiated cells: essentially defining a developmental program. This definition requires that gene expression states are heritable through mitotic cell divisions, but there is growing interest in understanding how gene expression states are heritable through meiotic cell divisions. We are using epigenomic approaches to profile diverse plant populations that have been collected over generational time to identify epialleles and to understand mechanisms important for stable transmission of these epiallelic states. Upon identification of epialles from populations over generational time we aim to explore the mechanisms that are essential to maintaining epiallelic states. The above figure demonstrates one possible mechanism that can lead to an epiallelic state. In this scenario, the maintenance methyltransferase MET1 fails to accurately maintain CG methylation which results in a breakdown of information flow between generations.

Population Epigenomics

Stacks Image 57
The lab is using a variety of epigenomic approaches in combination with diverse populations of plants to understand the impact of natural DNA methylation variation on phenotypic variation using three major approaches:

1. methylQTL
Genome-wide association studies (GWAS) is a powerful approach for identifying the genetic basis for phenotypic variation within large populations. We are interested in understanding how epialleles for in natural populations and to accomplish this goal we have used GWAS to map causal loci for variation in DNA methylation which we refer to as methylQTL. Understanding the types of genetic variants that can give rise to methylation variants is a major focus of our lab.

2. EpiGWAS
EpiGWAS is a variation of traditional GWAS that uses methylation polymorphisms as markers to test for association with a particular phenotype as opposed to the traditional approach which utilizes DNA polymorphisms. We are applying EpiGWAS to populations of plants to determine how much phenotypic variation results from DNA methylation polymorphisms.

3. Comparative Epigenomics
We are using comparative epigenomic approaches to understand both the variable and also the invariable portions of epigenomes by profiling DNA methylomes, histone tail modifications and RNAs from a variety of flowering plant species. Understanding why certain genes and gene families are hypervariably targeted by DNA methylation is important for revealing common functions in key biological processes.