Today begins with the session Systems biology & functional genomics. Here are some notes from those I was really interested into. The abstracts for all the talks can be found on the EMBO meeting dedicated page.
Genomic wide discovery of human enhancer
Eddy Rubin (JGI & LBNL, US)
The major part of the human genome is annotated as non coding. How are we supposed to understand the implication of a genomic region in a disease while we do not know what it does?
Enhancers are essential for gene functions; therefore, two things need to be determined: enhancers’ locations and genes these enhancers regulate. Two approaches are adopted: comparative genomics (blind approach) and ChIP-seq.
Comparative genomic enhancer discovery: we can compare two mammalian genomes, e.g. mouse and human. We can make a construct carrying a putative enhancer sequence and a LacZ reporter and transfect fertilized mice eggs: 3 mice embryos are followed for expression of the reporter. This is tedious and time-consuming, but nevertheless, more than 3,000 enhancer candidates have been tested. All this data is in a database called Enhancer browser. But this method has some limitations:
- this kind of discovery is based on sequence similarity between the two genomes having been compared. But there is no ifnormation about when or where enchancer is active;
- this method does not identiy enhancers that are not evolutionary conserved.
We come to the ChIP-seq which is a targeted discovery. Thanks to Next-Gen Sequencing (NGS), a very fine resolution is achieved; the reads are then mapped back to the genome. An example is discussed: p300. From ChIP-seq data, it comes out that a very small proportion of sequences identified as enhancers are under a high evolutionary constraint (~2%). Does ChIP-seq predict tissue-specific enhancer activiity? In this case, between 73 and 80% of the enhancers identified are active in the tissue they were found. Thus, p300 occupancy accurately identifies active enhancers with respect to the tissue. ~1,000 enhancers are discovered in this study and 1-2% of them only show high evolutionary constraints. It appears that a huge proportion of heart enhancers show no evolutionary constraint. Moreover, it does not make a difference whether evolutionary constraint is strong or not for a region to be enhancer:what is important is the p300 epigenomic mark. In conclusion, what is obvious is that enhancer conservation levels differ between tissues and many heart enhancers aer weakly conserved, i.e. hidden from comparative genomics.
For further insights of this study, check ChIP-Seq identification of weakly conserved heart enhancers (Nature Genetics 2010, doi:10.1038/ng.650)
Synthetic biology: from modules to systems
Wendell A. Lim (LIM lab, UCSF, US)
The hype cycle of new technologies: promises and perils. Where light we end-up? The following major points are to be adressed:
- applications: using living organisms for new production strategies;
- knowledge: critical approach to transform biology into foundational, principle-based science;
- tools: oowerful new ways to manipulate biology;
- education: new ways to teach biology.
Knowledge: how can synthetic biology transform our fundamental understanding of biology? Our current way is observe and analyze, afterwards correlate. What may be is principles-based and predictive way of adressing questions. In other words: going from what exists in nature (what we know about) to what may be and does not exist for now. For example,synthetic chemistry being a combination of principles and application defines a new philosophy. We need synthetic approaches to understand organizational logic of complex biological systems.
How can synthetic biology be used as a tool for deeper understanding? Example: how do biological systems achieve adaptation? There is an input, cells respond but automatically reset themselves; an output is produced. There are then 2 approaches to understand circuit functions and topodlogy: either study an existing system or create a new one. A computational enumerating of solutions for adaptation has been done and even if more thant 16,000 possible architectures appear and more than 400 f them are robust, all solutions map to only 2 families: there are therefore finite solutions. Thus, may we go towards biological design principles? The latter would govern general solutions that evolution converges on and serve as design guide for building diverse synthetic systems or for “reparing” diseased systems.
Tools: what are the new ways to perturb/manipulate biology? For example, it would be extremely useful to have generic way of using light to control interactions: this will constitute new ways of interrogating pathways (elucidte feedback architecture, spatial patterns, etc.).
Education: Synthetic biology is very compelling for young people; building is fun and engaging because it is not based on memorizaton of “how things are” and allows students to easily see the logci of how things work. The iGEM competition is a brilliant example of how undergrads can do science.