Two sides of a coin: chromatin-mediated packaging and regulation of metazoan genomes

Add event to my Google calendar Add event to my Google calendar Share this event on facebook E-mail this event
Wednesday, October 04, 2017 4:00pm - 5:00pm
Watson CIT - SWIG Boardroom (CIT241)

Two sides of a coin: chromatin-mediated packaging and
regulation of metazoan genomes

Michael Tolstorukov
Dept. of Molecular Biology
Mass General Hospital and Harvard Medical School

In metazoans, genomic DNA is packaged into chromatin, which
is composed of nucleic acids, histones, and other chromatin-associated
proteins. In addition to genome packaging, chromatin plays a major role in
controlling accessibility of DNA to the transcription machinery, i.e. in the
regulation of genome readout. Disruptions of this ‘epigenetic’ mechanism are
linked to many diseases and developmental disorders in humans.

Analysis of chromatin accessibility on a genome-wide scale requires
a combination of experimental, statistical, and computer science approaches. Together
with our experimental collaborators, my computational biology team has
developed a new methodology to measure chromatin accessibility, which
substantially expands the capabilities of existing methods. In my talk I will describe
this methodology and its application for the investigation of chromatin
regulation during development of cellular stress response. I will also discuss
using machine learning algorithms to generate comprehensive predictive models
of chromatin state based on the DNA accessibility data, which creates broad
opportunities for using our methodology in both basic and clinical research.

Plasticity of chromatin structure is modulated by altering occupancy
of chromatin factors on the genomic DNA. Among these factors, a special class
of proteins, called replication-independent histone variants, plays a critical
role in multiple cellular processes from chromosome segregation to
transcription regulation. Mutations in the histone variants are linked to
multiple human cancers, including pediatric brain and bone tumors. Despite
their importance, the regulation and evolution of the genes encoding the
histone variants themselves is poorly understood. I will describe the efforts
ongoing in my group to fill this gap as well as the potential biomedical
applications of our findings.