Council enjoys epigenetics and bioinformatics talks
By Ernie Hood
Like the icing on the cake, the two scientific presentations scheduled into most Council meetings provide members with some relief from making the tough decisions on grants and weighing in on budgetary matters. The winter National Advisory Environmental Health Sciences Council meeting was no exception, as members and other attendees were treated to a talk Feb. 20 on epigenetics, by a long-time NIEHS grantee, as well as a presentation on bioinformatics innovations being developed by an intramural researcher.
Challenging conventional wisdom about epigenetics
First up was Johns Hopkins University researcher and NIH Pioneer Award winner Andrew Feinberg, M.D., who described his work on “Investigating Epigenetic Plasticity in Development and in Response to the Environment.” By developing an integration of new conceptual, technological, epidemiological, and statistical approaches, Feinberg’s group is working to advance understanding of the role of epigenetics in areas ranging from cancer to common disease.
Feinberg began by saying epigenetics is where genes meet the environment. “The epigenome can integrate information that comes from your environment, from your genotypes, and the genotypes themselves can have an effect on the epigenome, so you have this complex interplay,” he explained. By adapting some of the statistical methods employed by astronomers, his group at the Johns Hopkins Center for Epigenetics has developed new tools to help understand the epigenetic contribution to diseases, such as cancer and neuropsychiatric disorders.
One of these new tools, comprehensive high-throughput arrays for relative methylation (CHARM), allows an unbiased look at the whole spectrum of epigenetic marks within the entire genome. Using CHARM to measure up to four million DNA methylation sites throughout the genome, the team discovered that, contrary to prior assumptions, most variable DNA methylation is not in islands, but in nearby sequences they termed shores. They discovered that aberrant methylation in cancer involves roughly equal gains and losses of DNA methylation at these shores, and involves much the same sequence involved in normal differentiation of widely disparate tissues.
Feinberg also shared his ideas regarding a non-Lamarckian model for the role of epigenetics in evolution, involving stochastic epigenetic variation as a driving force of development and evolutionary adaptation. He said the inherited stochastic variation model would provide a mechanism to explain an epigenetic role in selectable phenotypic variation, as well as the largely unexplained heritable genetic variation underlying common, complex diseases.
Bringing bioinformatics to the rescue
Leping Li, Ph.D., is the most recent NIEHS researcher to be awarded tenure, and is making great strides in bioinformatics, the essential tool for effectively processing and analyzing the huge amounts of data generated in contemporary biomedical research. In a presentation titled “Bioinformatics in Environmental Health Research,” Li described his work developing new bioinformatic tools and methods designed specifically to address questions related to environmental factors. For example, he and his group have developed novel tools for detection and analysis of transcription factor binding sites in DNA sequences. These motif analysis tools apply to large-scale, high-throughput DNA sequencing technologies, such as ChIP-seq. One, known as GADEM, performs motif discovery and identification, while another, known as coMotif, identifies primary motifs and potential coregulatory motifs in ChIP-seq data. Such computational analysis methods facilitate new, unbiased discovery and hypothesis generation.
Li described how the tools have been used to yield important new insights regarding the functions of estrogen receptor alpha in the uterus, which are critical for establishment and maintenance of pregnancy and are subject to environmentally influenced perturbations.
The motif analysis tools have been made available free to the public.
Li and his colleagues are also developing statistical and computational methods to identify differentially expressed isoforms from mRNA-seq data. As he explained, tools for detecting differential splicing could have a major impact in toxicogenomics, as examples exist where changes or imbalances in isoforms have been implicated in tumor development.
(Ernie Hood is a contract writer for the NIEHS Office of Communications and Public Liaison.)