Bejerano Lab, Stanford University

Understanding How the Human Genome Encodes Cis-Regulatory Function

Our lab aims to understand the cis-regulatory architecture of the human genome. To this end, we are building a unique suite of computational tools. We compile our own transcription factor binding site motif library, discover motifs de novo from sequence, perform genome wide binding site predictions, filter these using multiple species conservation, and more.

By massively running our tools on our large computer cluster, we start to decipher the transcription factor combinations that dictate promoter and enhancer specificity. We then employ functional genomics in the mouse and zebrafish model organisms to validate and improve our predictions.

We have also recently published and released GREAT (the Genomic Regions Enrichment of Annotations Tool), the first tool dedicated to the analysis of genome-wide cis-regulatory data. We have shown that GREAT can associate a set of genomic regions, such as ChIP-Seq peaks, with the nearby genes they likely regulate and perform enrichment analysis over many ontologies to find subsets of the genomic regions likely working in concert in well defined contexts (McLean et al., 2010). For more details see http://great.stanford.edu/.

Genomics of Morphogenesis and Differentiation in Development and Disease

One of the key mysteries of the cis-regulatory architecture of the human genome lies in its ability to encode the use and reuse of the same gene repertoire to build and maintain very different tissues and organs.

The cis regulatory regions underlying development also play significant roles in human disease onset and progression. For example, more and more cases are being discovered where an important developmental gene is intact, but a key regulatory region controlling it is mutated, triggering tissue specific disease pathology.

Our lab is focusing on studying the cis regulatory underpinnings of three processes in human development:

Limb morphogenesis - One of the first organs where cis-regulatory mutations have been clearly implicated in human congenital malformations. Having evolved from fins, limb regulatory regions can be rapidly screened for in zebrafish.
Placentation - The placenta is an understudied organ that has arisen in the mammalian clade. Misregulation of genes during placentation can lead to a number of disorders, including preeclampsia, premature birth, and other common birth defects.
Forebrain development - Improper development of the forebrain is thought to be involved in cognitive disorders such as autism and schizophrenia. The forebrain has also dramatically evolved in very recent human evolution.

Much of the information about the development of these tissues comes from the mouse model organism. We use ChIP-Seq, FAIRE-Seq, transfections and more, combined with our unique computational tools to expand our understanding of the gene networks underlying early organ development, as well as identify the genomic regions that control them.

Human Genome Evolution and Evolutionary Developmental Biology

Dobzhansky's famous quote, that "Nothing in biology makes sense, except in the light of evolution," is especially pragmatic in the genomic era. With over forty primate, mammalian and vertebrate genomes fully sequenced, and technological advances ushering in a flood of additional genomes, we are able to further interpret biological phenomena through the fascinating lens of time.

We ask questions of genome evolution, such as what are all the conserved regions in the human genome (Siepel et al., 2005) and what are their functions (Bejerano et al., 2005), which conserved regions display remote homologies due to shared origins (Bejerano et al., 2004a), how mobile elements can be co-opted to become conserved regulatory elements (Bejerano et al., 2006; Lowe et al., 2007; Lowe et al., 2010), and how conserved element loss affects organism fitness (McLean & Bejerano, 2008).

We are also fascinated with Evolutionary Developmental Biology, and together with David Kingsley's lab at Stanford we investigate regulatory aspects of vertebrate, mammalian and human specific evolution, looking to correlate sequence changes with trait evolution.

Ultraconserved Elements

Our early exploration of the human genome's non-coding landscape led to the surprising discovery of ultraconserved elements in the human genome (Bejerano et al., 2004b). Ultraconservation is the genomic phenomenon whereby certain genomic regions, most of them non-coding, are perfectly conserved across many mammals (and often beyond) for no clear reason. Adding to the mystery, knock-out mice have been produced for four different ultraconserved elements, with no clear phenotypic effect. However, we have shown that these sequences evolve under extreme negative selection in human (Katzman et al., 2007) and are practically never lost in naturally evolving mammals (McLean & Bejerano, 2008), strongly suggesting a fitness toll that has yet to be discovered. Many of the ultraconserved elements have been shown to work as developmental enhancers, and as such form the tip of the iceberg for our focus in understanding genomic cis-regulation.