MRC Functional Genomics Unit
University of Oxford
Topic: Mammalian Genomes
This afternoon's session started off with a topic I'm pretty far from, human and mammalian genomics. I've worked hard in my career to stay away from diploid genetics--I like the simplicity of viruses and prokaryotes by comparison. Today's talk was in 3 parts focusing on the human and mouse genomes, functional DNA and transcript maps and the future.
When you think about the word, genomics quite an all-encompassing concept involving sequences (obviously), comparisons, mapping, and new technologies. The word was actually coined around 1986 by Dr. Tom Roderick and you can read an interesting article article which is unfortunately not freely available. PubMed says there's a 'free final text' but they lied...I clicked the link and there was an error. If you come across the article and feel like sharing it, please link it in the comments!
Kuska, B. 1998. Beer, Bethesda and Biology. Journal of the National Cancer Inst. 90: 93.
Dr. Ponting went on to discuss the timeline of genomics with the Era of Genomics starting in 1999 and going into 2005 before it transitioned into focusing on evolutionary genomics which has run us up to 2010 and now things have been focusing on disease genomics. I found this interesting as I entered the field in 2003 during the era of genomics and our lab worked with collaborators at The Inst. for Genomic Research (TIGR) for our sequencing projects; we jumped into ecology and evolution right away which was prime time for the transition on the timeline and oddly enough where am I now? Working in disease genomics--granted not from the human perspective, but still. It was interesting to see the timeline follow the trajectory of my career thus far.
The best thing about the era of genomics was that there were no preconceived notions of what a genome was, how big it was, the composition etc...so 'anything was possible.'
In 2001, Nature and Science published papers highlighting draft human genomes from two competing groups.
Lander et al., 2001. Initial sequencing and analysis of the human genome. Nature. 409: 860.
Science magazine issue 2001: Human Genome Special Issue
Dr. Ponting was not involved in the Science magazine article, but was involved with the Nature article. We'll bullet the main points out of the Nature article and go head and click over to the Science issue if you'd like to see what was published by Science. Chris Pontings slides will be available on the evomics website.
- 96% of the euchromatic genome sequenced (94% of total human genome)
- They went from 10% to 90% coverage of the genome in 15 mos.
- Sequence data was to be made available without restriction as soon as possible.
- Transposable elements dominated, 45% (and may be more), so of course this begged the question why are they there and what do they do?
Further research confirmed:
- The human genome landscape was more complicated that in invertebrates and secreted products were more complex.
- ~20,000 genes are protein encoding
- Horizontal gene transfer into the human genome was actually quite rare.
- Large chunks of the genome were duplicated, many times more than once and from one chromosome to another.
- Interestingly, most mutation occurs in males than females (nearly 2 times higher in males).
- Increased recombination was found at the end of chromosomes and in shorter chromosomes as opposed to in the middle or in longer chromosomes.
In 2004, they finished the euchromatic genome and of course it generated a lot of questions:
- How many genes?
- How much of the DNA is functional?
- How do isochores arise?--Isochores are large regions of DNA that are homogenous in GC content.
- How is recombination controlled?
- How does our genome compare with others?
- Are transposon derived sequences functional?
- What makes us, us?
- How are transcription and splicing controlled?
Truly...how useful is the human genome?
Ever wonder 'who' the reference human genome is? Volunteers from Buffalo, New York--so now you know!
From Dr. Ponting and colleagues studies of genome comparison between mouse and human they developed a three state model fo mammalian genomes (Ponting, CP; C Nellaker, and S Meader. 2011. Annu Rev Genomics Human Genet).
Some take home messages from all the human genome and mouse genome comparison work were as follows:
- Conservation of genes does not equal functional conservation.
- Constrained genomic data also does not equal function conservation necessarily.
- Positive selection does not equal adaptive evolution.
- Effective population size is of great importance. A question that was asked with respect to humans is what effective population to use? Modern or ancient because dynamics working on each population would be very different. The answer provided was that ideally it would be the harmonic mean between modern and ancient effective population sizes.
This section had to do a great deal with the ENCODE project. One of the ENCODE papers asserted that 80% of the genome is functional meaning it was linked to a biochemical process; whereas Dr. Ponting contends you cannot define functionality in this manner and that in reality there is only about 10% of the genome that could be considered functional. And the debate in the literature continues.
The ENCODE project gave us a great deal of data and analysis. 5 Terabytes of data and 3,010 experiments! Our understanding of annotation, function, analyses that assist regulatory element study (ChIP-seq, RNA-seq, DNase-Seq) were developed and genome wide association studies were being started to determine disease associated variants within the human genome. They found that 88.1% of traits that could lead to disease are 'laid down earlier in life, as opposed to later'. And they are not so deleterious as to impede propagation so these traits are passed on. Correlating disease variants with cell-types was also studied.
When transcriptomes were being studied, there was a lot of RNA expression going on at low levels and not by protein encoding genes. They found 31% of bases in sequenced transcripts were intergenic (not in genes). But this topic is under debate, they call it 'dark matter transcription'...
And there have been several articles debating this topic:
Phillips ML. 2010. The abundance of transcripts from the genome may have been overestimated. Nature News.
Johnson et al., 2006. Dark matter in the genome: evidence of widespread transcription detected by microarray tiling experiments. Trends in Genetics.
So where do we go from here?
- Studies should move from interspecies to intraspecies genome comparisons; such as looking at differences among human populations.
- Studies should move toward population genetics.
- How do we locate variants that underlie genetic traits?
- How do we find variants that differentiate species?
- Phenotyping the human
- Increased understanding of cell types
And the list goes on!
Currently the pace of change in genomics is constant; which is frustrating but also exciting, the goal is to keep pace and keep the questions coming.
"The most important commodity in genomics is ideas" ~ C. Ponting, 2013.
If you are interested in a postdoc or fellowship and would like to work at Oxford, contact Dr. Ponting at email@example.com