Our group is particularly interested in strategies to assess epigenetic variation within and between cell populations, and to reveal transcription factors and gene networks controlling epigenetic dynamics during normal development and diseases. The term “epigenetic” mainly refers to histone modifications and DNA methylation, which are alternative ways to control gene expression while maintaining the nucleotide sequence of the genome. In other words, epigenetic changes allow cells with identical genomic content to demonstrate distinct phenotypes.
Our early studies were focused on the analyses of the genomics, transcriptomics, and epigenomics of developing brains and brain tumors. We exploited genome-wide techniques, microarrays, and high-throughput sequencing to uncover gene/protein networks with consistently compromised functions in brain tumors. In addition to the identification of diagnostic/prognostic molecular markers, we addressed the following questions: 1) the extent and functional aspects of epigenetic variations in normal and diseased tissues; 2) the crosstalk between epigenetic and genetic variations regulating signaling pathways; and 3) the dynamic interactions of functional genomic domains, such as CpG islands and repeats, in the epigenomic configuration.
Recently, our group has implemented genome-wide hairpin bisulfite sequencing to assess asymmetric DNA methylation between the double strands of DNA. This approach enables the assessment of the fidelity of DNA methylation inheritance. Since the challenges in epigenomic studies are not limited to data generation but also reside in interpretation of such enormous amounts of data, we have developed computational methods to decode bisulfite sequencing data: 1) the analysis of “methylation entropy” to quantitatively assess the variation of DNA methylation patterns in a given cell population; 2) nonparametric Bayesian clustering to detect bipolar methylated genomic loci; and 3) a computational pipeline to decipher the heterogeneity in DNA methylation patterns and to infer cell-type specific methylated loci.
To understand genome-environment interactions with our collaborators, we have explored epigenomes to identify aberrations associated with various human diseases. We have also made the effort to define the individual variation and longitudinal pattern of the epigenome during early postnatal human development. We are particularly interested in understanding the environmental influence on the transcription factor networks controlling cell specification. For inflammatory diseases, we have determined the epigenetic changes on genes coding for transcription factors including Wilms tumor-1 (WT1) and homeobox genes, which have been shown to be critical for normal tissue development and cell specification. Our study on pediatric patients may shed light on how environmental factors may alter the normal developmental process via these critical factors.
DMEAS is the first user-friendly tool dedicated to analyze methylation entropy for the quantification of epigenetic variation. The DMEAS program, user guide, and all the testing data are freely available.
HBS analyzer is the first command-line-based open-source tool to process genome-wide hairpin bisulfite sequencing data. It accepts Illumina paired-end sequencing reads as input, performs alignment to recover the original (pre-bisulfite-converted) DNA sequences, and calls methylation status for cytosines on both DNA strands. The HBS analyzer program, user guide, and all the testing data are freely available.
|Sharmi Banerjee||Visiting Student|
|Megan Harrigan||BREU Student|
|Jianlin He||Research Associate|
|Emma McCoig||BREU Student|
|Alexander Murray||Graduate Research Assistant|
|Keaton Solo||BREU Student|
|Zhixiong Sun||Visiting Student|
|Xiguang Xu||Visiting Student|