Discussion Board

Topic: Nature essay

From: Alex Tolley
Location: Los Gatos, CA
Date: 01/26/2007

Below is an essay from the current issue of nature, suggestion a potential revolution in how we think of organisms and evolution. I think it goes further than the consequences of horizontal gene transmission. As you have alluded to in Darwins Radi/Children, Margulies in Aquiring Genomes, we can see that the potential effects of genes on organism evolution via non-genealogical means may be more common than previously thought. I personally see it as more of a continuum:

1. Standard, direct inheritance of genes.
2. Horizontal transmission in microbes, retroviruses etc.
3. Eucaryotes as permanent collections of genomes. (rare event in life's hsitory - but if Margulies is correct, not so rare).
4. "Diseases" as impactors on host phenotype and behavior(genome interactions but not permanent)
5. Co-evolution of symbiotes, commensals and organisms in ecological niches. (system interaction of separate genomes).


Nature 445, 369 (25 January 2007) | doi:10.1038/445369a; Published online 24 January 2007
ConnectionsBiology's next revolution

Nigel Goldenfeld1 and Carl Woese2

1. Nigel Goldenfeld is in the Department of Physics and Institute for Genomic Biology, University of Illinois at Urbana-Champaign, 1110 West Green Street, Urbana, Illinois 61801, USA.
2. Carl Woese is in the Department of Microbiology and Institute for Genomic Biology, 601 South Goodwin Avenue, Urbana, Illinois 61801, USA.

Top of page

The emerging picture of microbes as gene-swapping collectives demands a revision of such concepts as organism, species and evolution itself.

One of the most fundamental patterns of scientific discovery is the revolution in thought that accompanies a new body of data. Satellite-based astronomy has, during the past decade, overthrown our most cherished ideas of cosmology, especially those relating to the size, dynamics and composition of the Universe.
Biology's next revolution


Similarly, the convergence of fresh theoretical ideas in evolution and the coming avalanche of genomic data will profoundly alter our understanding of the biosphere  and is likely to lead to revision of concepts such as species, organism and evolution. Here we explain why we foresee such a dramatic transformation, and why we believe the molecular reductionism that dominated twentieth-century biology will be superseded by an interdisciplinary approach that embraces collective phenomena.

The place to start is horizontal gene transfer (HGT), the non-genealogical transfer of genetic material from one organism to another  such as from one bacterium to another or from viruses to bacteria. Among microbes, HGT is pervasive and powerful  for example, in accelerating the spread of antibiotic resistance. Owing to HGT, it is not a good approximation to regard microbes as organisms dominated by individual characteristics. In fact, their communications by genetic or quorum-sensing channels indicate that microbial behaviour must be understood as predominantly cooperative.

In the wild, microbes form communities, invade biochemical niches and partake in biogeochemical cycles. The available studies strongly indicate that microbes absorb and discard genes as needed, in response to their environment. Rather than discrete genomes, we see a continuum of genomic possibilities, which casts doubt on the validity of the concept of a 'species' when extended into the microbial realm. The uselessness of the species concept is inherent in the recent forays into metagenomics  the study of genomes recovered from natural samples as opposed to clonal cultures. For example, studies of the spatial distribution of rhodopsin genes in marine microbes suggest such genes are 'cosmopolitan', wandering among bacteria (or archaea) as environmental pressures dictate.

Equally exciting is the realization that viruses have a fundamental role in the biosphere, in both immediate and long-term evolutionary senses. Recent work suggests that viruses are an important repository and memory of a community's genetic information, contributing to the system's evolutionary dynamics and stability. This is hinted at, for example, by prophage induction, in which viruses latent in cells can become activated by environmental influences. The ensuing destruction of the cell and viral replication is a potent mechanism for the dispersal of host and viral genes.

It is becoming clear that microorganisms have a remarkable ability to reconstruct their genomes in the face of dire environmental stresses, and that in some cases their collective interactions with viruses may be crucial to this. In such a situation, how valid is the very concept of an organism in isolation? It seems that there is a continuity of energy flux and informational transfer from the genome up through cells, community, virosphere and environment. We would go so far as to suggest that a defining characteristic of life is the strong dependency on flux from the environment  be it of energy, chemicals, metabolites or genes.

Nowhere are the implications of collective phenomena, mediated by HGT, so pervasive and important as in evolution. A computer scientist might term the cell's translational apparatus (used to convert genetic information to proteins) an 'operating system', by which all innovation is communicated and realized. The fundamental role of translation, represented in particular by the genetic code, is shown by the clearly documented optimization of the code. Its special role in any form of life leads to the striking prediction that early life evolved in a lamarckian way, with vertical descent marginalized by the more powerful early forms of HGT.

Refinement through the horizontal sharing of genetic innovations would have triggered an explosion of genetic novelty, until the level of complexity required a transition to the current era of vertical evolution. Thus, we regard as regrettable the conventional concatenation of Darwin's name with evolution, because other modalities must also be considered.

This is an extraordinary time for biology, because the perspective we have indicated places biology within a context that must necessarily engage other disciplines more strongly aware of the importance of collective phenomena. Questions suggested by the generic energy, information and gene flows to which we have alluded will probably require resolution in the spirit of statistical mechanics and dynamical systems theory. In time, the current approach of post-hoc modelling will be replaced by interplay between quantitative prediction and experimental test, nowadays more characteristic of the physical sciences.

Sometimes, language expresses ignorance rather than knowledge, as in the case of the word 'prokaryote', now superseded by the terms archaea and bacteria. We foresee that in biology, new concepts will require a new language, grounded in mathematics and the discoveries emerging from the data we have highlighted. During an earlier revolution, Antoine Lavoisier observed that scientific progress, like evolution, must overcome a challenge of communication: "We cannot improve the language of any science without at the same time improving the science itself; neither can we, on the other hand, improve a science without improving the language or nomenclature which belongs to it." Biology is about to meet this challenge.

Further reading

Frigaard, N., Martinez, A., Mincer, T. & DeLong, E. Nature 439, 847850 (2006).

Sullivan, M. et al. PLoS Biol. 4, e234 (2006).

Pedulla, M. et al. Cell 113, 171182 (2003).

Vetsigian, K., Woese, C. & Goldenfeld, N. Proc. Natl Acad. Sci. USA 103, 1069610701 (2006).

Re: Nature essay

From: Greg Bear
Date: 01/26/2007

Wow! This certainly lays down the gauntlet. These gentlemen have the track record to justify making such strong assertions--and the stamina to fight until they are widely accepted. Bravo to them.

What's astonishing to me is how far biology has come since I wrote DARWIN'S RADIO. I should also point out the big contributions made by others to our thinking: Eshel Ben Jacob's ground-breaking work on bacterial colonies under stress, Lynn Caporale's prescient seminar on evolution... Bravo to all!

And now for the next big step--when we realize what Barbara McClintock actually meant when she described the genome as an ecosystem.

The same patterns that govern neural activity in the brain may actually guide and describe the interactive and reactive nature of living things from top to bottom.

Re: Nature essay

From: Alex Tolley
Location: Los Gatos, CA
Date: 01/26/2007

I think it is more than McClintock's ecosystem concept: I think you have touched on it with your analogy to neural processing. Evolution by natural selection (IMO the only robust theory in biology) ultimately requires physical replicators - genes. We have tended to focus on the details of those genes: structure, generated phenotypes, etc., but biologists have tended not to see genomes as an abstraction for processing (although see Kauffman, The Origins of Order). We shouldn't be so surprised when we see different patterns of interaction and processing that generates the necessary order for signal processing. Your neural analogy is helpful here - some possible, but loose, analogies in information processing patterns in brains and genomes:

1. Spreading activation waves : transcription cascade
2. Hebbian weight pattern : pattern of transcription enhancer sites
3. Minsky's Society of Mind : gene interaction
4. Meme acquisition/learning by instruction : horizontal gene transmission
5. Cultural evolution : genome ecosystem (?)

Now here is where I think it gets interesting. There are a number of different software algorithms to optimize systems, some modeled on living systems (genetic algorithms) some not (linear programming). Some are similar-ish (metropolis/simulated annealing ~: sex (?)). What is relevant is that these algorithms all have performance vs cost envelopes. Genetic algorithms are very powerful, but very expensive, whereas linear programming is limited but fast. Given that evolution by natural selection favors approaches that maximize this trade off in different scenarios, it would not surprise me to learn that living systems use these different algorithmic strategies in their different information processing manifestations - genomes, brains; algorithms that at their core are the same in both systems, but manifested differently depending on the actual substrate.

To be really speculative, one could start to think about how living systems could interact with incompatible systems, much like we design interactions between different software languages. The simplest is the equivalent of a System.exec function - genes -> genes, or mitochondrial, chloroplast and nucleosome genome interactions. More interesting is the use of APIs of different types - ribosomes, protein transcription factors (?). Current thinking is that terrestrial and exo-biologies are probably completely incompatible and thus almost like separate, non-interacting systems, except at the macro level. But is this true? Could living systems use any of the novel genome interacting strategies to enhance fitness? The staple of much scfi/horror I know, but maybe not so impossible...?

Re: Nature essay

From: Greg Bear
Date: 02/02/2007

Ah--can the Thing really be so Thingy? Good question, my guess is, probably not. (In my view, an ecosystem also functions as an interacting network--and can solve problems on its own level by, for example, recruiting and adapting or causing species to adapt.)

Re: Nature essay

From: Alex Tolley
Location: Los Gatos
Date: 02/07/2007

"can the Thing really be so Thingy?" - did I get the Clarke "You really squiggled my squiggle" response :-

I hadn't quite thought about ecosystems as being so dynamic and therefore an algorithmic substrate. Food for thought.
I suppose if we think of the node as a species and a connection as the impact on another species, that would map quite well to a Hebbian network in the static case, and the node/connection replacement as a GA or similar in the dynamic case.

Re: Nature essay

From: Greg Bear
Date: 02/07/2007

As you say, Alex... I'm not even past Bayesian, myself!

Re: Nature essay

From: Chris
Location: Huntington Beach
Date: 01/16/2008

I'm new to the site, and absoultely love the speculation in these books. I'm not sure if it's been posited before, but I'm curious as to whether anyone sees DNA and RNA as a base-four computer assembly language? A, C, G, and T vs. 0 and 1. If this is the case it adds significantly to the complexity of trying to figure out what genes code for.

Re: Nature essay

From: Greg Bear
Date: 01/18/2008

There's been speculation about DNA "grammar," and even linguistic analysis of DNA sequences, and some sort of computational analysis is probably part of the big picture, but that picture grows increasingly complicated. Pure computational analysis is likely inadequate, but the grammatical part still seems to have some power. The genes are like keys on a piano, or spices in a spice rack (as my wife compares them). Whether you play the keys, how many times you play them, and how long you sustain the note... All that seems to play a very major role. So--is DNA essentially a musical score, and life the composition?

Re: Nature essay

From: Alex Tolley
Location: Los Gatos
Date: 01/18/2008

GB: "is DNA essentially a musical score, and life the composition?"

Nice metaphor, but I don't think so, simply because music is a linear sequence, whist the chords represented by DNA transcription and translation are interacting.

I think the computer code metaphor is better, although probably not accurate either. I am fascinated by the recent work showing that many "genes" are not translated to proteins but the rna transcripts are functional, much as they must have been in the early "rna world".

There is also the intriguing, but unproven idea, based on recent work that shows some DNA/RNA sequences do not exist. The suggestion is that they may be "lethal" sequences. Nice idea for a story, but I have no idea if it is real or not, although one should be able to test it readily enough by inserting the sequences and seeing what happens.

Re: Nature essay

From: Greg Bear
Date: 01/20/2008

True enough--so imagine a million pianos playing a million different tunes together, yet they all add up to something that works... Software is certainly a workable metaphor, one that I've used for years (going back to asserting that DNA was computational, back in the 1980s).

Respond to this discussion

GregBear.com is currently being updated and new comments to the discussion board have been paused until the new site is ready on September 6, 2015. Please check back then!

See Also...

Archives: [Oct-Dec 2004] [Jan-June 2005] [July-Dec 2005] [Jan-June 2006] [July 2006] [Aug-Dec 2006] [2007] [2008] [2009] [2010] [2011] [2012] [2013] [2014] [2015] [Current] [Search Blog Archives]