Mapped Out: A Complex Landscape of Genetic Interactions
What's Hot in Biology May/June 2011
by Jeremy Cherfas
A skyscape of white dots picks out a circle on a black background. Connecting the dots are faint wispy trails of webbing, and among the white are galactic clusters of blue, red, lime, teal, yellow, pink, and other lurid colors. But this is no night sky. The labels, colored like the galaxies they refer to, give that away: "Mitochondria," "Nuclear-cytoplasmic transport," "DNA replication & repair."
This picture—which would take considerably more than a thousand words to explore fully—is the centerpiece of the paper at #6 and shows, as the paper’s title has it, "the genetic landscape of a cell.
Charles Boone, at the Donnelly Centre for Cellular and Biomolecular Research of the University of Toronto in Canada, and a large team compiled a detailed look at the interrelationships between thousands of genes of yeast (Saccharomyces cerevisiae) into a picture of the genetic landscape. What is both astonishing and humdrum about it is that the functional clusters picked out in color were not preselected. They emerged from what the authors call a "functionally unbiased genetic interaction map."
In a sense, that is to be expected. One might assume that genes that operate in concert as part of a complex function would interact more often than genes not linked in this way. The surprise is how clearly this shows up in the landscape map.
Saccharomyces cerevisiae is a species of budding yeast. From
the Wiki Commons.
Yeast has about 6,000 genes. Boone and his group had systematically deleted single genes, and showed that very few—about 20%--are essential. The rest the organism can manage without, although it may take a fitness hit, reproducing more slowly than a normal yeast.
The team then turned its attention to double mutants; did they grow more or less slowly than one might expect based on the combined effect of the two mutations taken singly? A difference in the double mutant would indicate that the two genes interacted in some way. A pairwise screen of 1,712 genes, chosen at random, yielded about 5.4 million comparisons, among which were 170,000 significant interactions between pairs of genes.
Most were negative, where the absence of two genes had an even greater effect than expected from the absence of either one alone. Genes that are part of the same biological process tend to share similar genetic interactions, and this was the key to building the network of the landscape.
Each node on the network is a gene, its distance from other genes based on the similarity between its pattern of interactions and the overall pattern. Genes of known function clearly clustered together. Indeed, the team used the clusters to pin functions on previously uncharacterized genes.
Taking a broader perspective, most of the genes had very few interactions, while a small number were highly connected as network hubs. Among the hubs, there were some that were much more likely to have negative interactions than positive, and a smaller number that were more likely to be positive than negative.
"Genes that are part of the same biological process tend to share similar genetic interactions, and this was the key to building the network of the landscape..."
The two sets were functionally distinct. Negative interactions were common among genes responsible for the cell’s normal progress through the cycle of growth and cell division. This analysis thus confirms the notion of a series of checkpoints that keep cells healthy by blocking progress if anything is going wrong. Positive interactions were found among genes responsible for DNA translation and the processing of RNA which, the team says, "may suggest that defects in the translation machinery somehow mask phenotypes that would otherwise be expressed in normal cells."
Network hubs were interesting in other ways, too. There is a clear link between the degree to which a gene is connected and its impact on fitness as a single mutant. The more severe the fitness defect, the greater the number of positive and negative interactions. Hubs also exhibited pleiotropy; that is, they were connected to a greater number of distinct functional groups.
Looking across 23 different fungi related to yeast, network hub genes were more highly conserved than other genes. Zooming in, interactions among genes sometimes revealed bridges between different processes. Genes involved in secretion and vesicle transport were among the most highly connected, emphasizing their role as channels of communication.
These are just a few of the highlights from the paper, which is as rich as the landscape it pictures. The progress from studying the effect of single-gene mutations, to two-gene interactions, to the full constellation of interactions that defines the genetic landscape has confirmed in entirely new ways that cellular processes are highly organized and well buffered against disturbance.
Further advances are promised, for example in drug discovery and a deeper understanding of the complex routes from genotype to phenotype. One can only imagine the insights that will emerge from comparing the genetic landscapes of different organisms.
Dr. Jeremy Cherfas is Science Writer at Bioversity International, Rome, Italy.
What's Hot in Biology | |||
---|---|---|---|
Rank | Paper |
Cites This Period Nov-Dec 10 |
Rank Last Period Sep-Oct 10 |
1 | J.J. Qin, et al., "A human gut microbial gene catalogue established by metagenomic sequencing," Nature, 464(7285): 59-65, 4 March 2010. [14 institutions worldwide] *563GZ | 32 | 6 |
2 | R.C. Friedman, et al., "Most mammalian mRNAs are conserved targets of microRNAs," Genome Res., 19(1): 92-105, January 2009. [MIT, Cambridge, MA; Whitehead Inst., Cambridge, MA; Howard Hughes Med. Inst., Cambridge, MA] *390YQ | 31 | † |
3 | L.A. Hindorff, et al., "Potential etiologic and functional implications of genome-wide association loci for human diseases and traits," PNAS, 106(23): 9362-7, 9 June 2009. [NIH, Bethesda, MD] *456CN | 30 | † |
4 | D.E. Harrison, et al, "Rapamycin fed late in life extends lifespan in genetically heterogeneous mice," Nature, 460(7253): 392-5, 16 July 2009. [7 U.S. institutions] *470MO | 29 | 2 |
5 | S.B. Ng, et al., "Targeted capture and massively parallel sequencing of 12 human exomes," Nature, 461(7261): 272-6, 10 September 2009. [U. Washington, Howard Hughes Med. Inst., Seattle; Agilent Technologies, Santa Clara, CA] *492KN | 27 | 3 |
6 | M. Constanzo, et al., "The genetic landscape of a cell," Science, 327(5964): 425-31, 22 January 2010. [15 institutions worldwide] *546BS | 26 | † |
7 | Y. Tanaka, et al., "Genome-wide association of IL28B with response to pegylated interferon-a and ribavirin therapy for chronic hepatitis C," Nature Genetics, 41(10): 1105-9, October 2009. [17 Japanese institutions] *500UG | 25 | 7 |
8 | C. Choudhary, et al., "Lysine acetylation targets protein complexes and co-regulates major cellular functions," Science, 325(5942): 834-40, 14 August 2009. [Max Planck Inst. Biochem., Martinsried, Germany; U. Copenhagen, Denmark] *487AK | 24 | † |
9 | S. Geisler, et al., "PINK1/Parkin-mediated mitophagy is dependent on VDAC1 and p62/SQSTM1," Nature Cell Biol., 12(2): 119-31, February 2010. [U. Tubingen, Germany] *550PD | 23 | † |
10 | A. Nikolaev, et al., "APP binds DR6 to trigger axon pruning and neuron death via distinct caspases," Nature, 457(7232): 981-9, 19 February 2009. [Genentech, South San Francisco, CA; Salk Inst., La Jolla, CA] *408HF | 21 | † |
SOURCE: Thomson Reuters Hot Papers Database. Read the Legend. |