Title: Chromatin modifications, gene expression, and regulatory networks

Speaker: Raja Jothi, National Heart, Lung, and Blood Institute

Abstract

First, I will discuss our results on the connection between chromatin modifications and gene expression. 
We analyzed chromatin modifications of induced and repressed genes in T-cells upon T-cell receptor signaling. 
For a majority of inducible genes, activating chromatin modifications were already present at the promoter 
and in the gene body even while the genes were silent in the resting cells. Similarly, genes that were silenced 
upon T-cell activation retained activating chromatin modifications after being silenced. Our results indicate 
that chromatin modifications poise inducible mRNA and miRNA genes for expression even before the signal for expression arrive. 

Next, I will present SISSRs (Site Identification from Short Sequence Reads), a novel algorithm for precise identification 
of in vivo transcription factor binding sites from ChIP-Seq data. We applied SISSRs on ChIP-Seq data for three widely-studied 
and well-characterized human transcription factors---insulator binding protein CTCF, neuron-restrictive silencer factor NRSF 
(also known as REST, for repressor element-1 silencing transcription factor), and transcription activator protein STAT1---
to gain novel biological insights. In particular, we show for the first time CTCF demarcates the active and repressed regions 
of the chromatin on a genome-scale. 

Finally, I will show that regulatory proteins within a hierarchical framework have distinct dynamic properties. I will present 
a novel graph-theoretical algorithm, which we apply on yeast transcription regulatory network to identify three mutually exclusive 
layers of transcription factors: initiators, propagators and effectors. By overlaying diverse genomic datasets on the hierarchical
 structure, we show that regulatory proteins within a hierarchical framework have dynamic attributes that are consistent within a 
layer and distinct across layers, thus making the network more robust and adaptable. Our findings have direct implications 
in synthetic biology experiments aimed at engineering transcription networks.