Algorithmic view of historicity and separation of scales in biology

A Science publications is one of the best ways to launch your career, especially if it is based on your undergraduate work, part of which you carried out with makeshift equipment in your dorm! That is the story of Thomas M.S. Chang, who in 1956 started experiments (partially carried out in his residence room in McGill’s Douglas Hall) that lead to the creation of the first artificial cell (Chang, 1964). This was — in the words of the 1989 New Scientists — an “elegantly simple and intellectually ambitious” idea that “has grown into a dynamic field of biomedical research and development.” A field that promises to connect biology and computer science by physically realizing John von Neumann’s dream of a self-replication machine.

Read more of this post

Did group selection play a role in the evolution of plasmid endosymbiosis?

plasmidBacterial plasmids are nucleotide sequences floating in the cytoplasm of bacteria. These molecules replicate independently from the main chromosomal DNA and are not essential to the survival or replication of their host. Plasmids are thought to be part of the bacterial domain’s mobilome (for overview, see Siefert, 2009), a sort of genetic commonwealth which most, if not all, bacterial cells can pull from, incorporate and express. Plasmids can replicate inside a host and then move to another cell via horizontal genetic transfer (HGT), a term denoting various mechanism of incorporation of exogenous genetic material.
Read more of this post

Continuing our exploration of group selection

This Tuesday, I gave the second of two presentations for the EGT Reading group, both focused on the theory of group selection. Though I am currently working outside of academia, it has been a pleasure to pursue my interests in ecology, and our group discussions have proven to be both enjoyable and challenging.

The first presentation [pdf] is a review of a 2011 paper written by Marshall. It argues that when the models underlying inclusive fitness theory (IFT) and group selection are formally identical. However, as I tried to show during the presentation, this formal equivalency only holds for one specific type of group selection – group selection as the partitioning of selection between groups from selection within groups. It no longer holds when we consider the more restrictive definition of group selection as “natural selection on groups” in strict analogy to individual selection (this, incidentally, is the definition of group selection I gave in my last blog post)

Marshall J.A.R. (2011). Group selection and kin selection: formally equivalent approaches, Trends in Ecology & Evolution, 26 (7) 325-332. DOI:

The second presentation [pdf] is a review of a paper by Paulsson (2002). That paper presents an interesting case of multi-level (group) selection, where the “individuals” are plasmids – self-replicating gene clusters in the cytoplasm of procaryotes – and the “groups” are the plasmid-hosting cells. It’s a nice illustration of the basic dilemma that drives group selection. Inside a cell, plasmids which replicate faster have an advantage over their cell mates. But cells in which plasmids replicate too fast grow slower. Thus, at the level of individuals selfishness is favored, but at the level of groups altruism is favored. Paulsson’s paper explains the mechanisms of plasmid replication control; sketches up models of intra- and inter-cellular selection gradients; and explains how conflicts between individual- and group-selection are resolved by plasmids. He also considers a third level of selection on lineages, but both Artem and I were confused about what exactly Paulsson meant.

Paulsson, J. (2002). Multileveled selection on plasmid replication. Genetics, 161(4): 1373-1384.

Programming playground: Cells as (quantum) computers?

Nearly a year ago, the previous post in this series introduced a way for programmers to play around with biology: a model that simulated the dynamics of a whole cell at unprecedented levels of details. But what if you want to play with the real thing? Can you program a living cell? Can you compute with molecular biology?


Could this single-celled photosynthetic algae be your next computer?

Biology inspired computation can probably be traced back as far back as Turing’s (1948) introduction of B-Type neural networks. However, the molecular biology approach is much more recent with Adleman (1994) proposing DNA computing, and Păun (2000) introducing membrane computing with P-systems. These models caused a stir when they appeared due to the ease of misrepresenting their computational power. If you allow the cells or membranes to carry on exponential rate of reproduction for an arbitrarily long time, then these systems can solve NP-complete problems quickly. In fact, it is not hard to show that this model would allow you to solve PSPACE-complete problems. Of course, in any reasonable setting, your cells can only grow at an exponential rate until they reach the carrying capacity of the environment you are growing them in. If you take this into account then efficient DNA and membrane computing are no more powerful than the usual definition of efficient computation — polynomial time on a Turing machine.

The stirred (i.e. inviscid) nature of membrane and (early approaches to) DNA computing provide substantial constraints for empirical realizations, and scalability of bio-computing. In these early models, regulatory molecules are reused in the self-mixing environment of the cell, and gates correspond to chemical reactions. As such, gates are temporary; and the information carrying molecule must change at every step of the computation to avoid being confused with residue from the previous step. This made implementing some gates such as XNOR — output 1 only if both inputs are the same — experimentally impossible (Tamsir, 2011): how would you tell which input is which and how would the gate know it has received both inputs and not just an abnormally high concentration of the first?

To overcome this, Bonnet et al. (2013) designed a cellular computation model that more closely resembles the von Neumann architecture of the device you are reading this post on. In particular, they introduced a cellular analog of the transistor — the transcriptor. The whimsical name comes from the biology process they hijacked for computation, instead of electric current flowing on copper wires the researchers looked at the “transcriptional current” of RNA polymerase on DNA “wires”. Only if a control signal is present does the transcriptor allow RNA polymerase to flow through it; otherwise it blocks them, just like an electric transistor. By putting several transcriptors together, and choosing their control signals, Bonnet et al. (2013) can implement any logic gate (including the previously unrealized NXOR) just as an electrical engineer would with transistors. What matters most for connecting to quantum computing, is the ability to reliably amplify logical signals. With amplifying gates like AND, OR, and XOR, the authors were able to produce more than a 3-fold increase in control signal. For further details on the transcriptor listen to Drew Endy explain his group’s work:

Taking inspiration from biology is not restricted to classical computation. Vlatko Vedral provides a great summary of bio-inspired quantum computing; start from top down, figure out how biology uses quantum effects at room temperature and try to harness them for computation. The first step here, is to find a non-trivial example of quantum effects in use by a biological system. Conveniently, Engel et al. (2007) showed that photosynthesis provides such an example.

During photosynthesis, an incident photon becomes an ‘exciton’ that has to quickly walk through a maze of interconnected chlorophyll molecules to find a site where its energy can be used to phosphorylate used-up ADP into energy-carrying ATP. Unfortunately, if the exciton follows a classical random walk (i.e. spreads out in proportion to the square root of time) then it cannot reach a binding site before decaying. How does biology solve this? The exciton follows a quantum walk! (Rebentrost et al., 2009)

It is cool to know that we can observe a quantum walk, but can that be useful for computation? My former supervisor Andrew Childs (2009; see also Childs et al., 2013) is noted for showing that if we have control over the Hamiltonian defining our quantum walk then we can use the walk to do universal computation. Controlling the Hamiltonian generating a quantum walk is analogous to designing a graph for a classical walk. Theoretical work is still needed to bridge Rebentrost et al. and Childs, since (as Joe Fitzsimons pointed out on G+) the biological quantum walk is not coherent, and the decoherence that is present might doom any attempt at universal computation. The last ingredient that is needed is a classic controller.

Since the graph we need will depend on the specific problem instance we are trying to solve, we will need a classical computer to control the construction of the graph. This is where I hope synthetic biology results like Bonnet et al. (2013) will be useful. The transcriptors could be used as the classic control with which a problem instance is translated into a specific structure of chlorophyll molecules on which a quantum walk is carried out to do the hard part of the computation. The weak quantum signal from this walk can then be measured by the transcriptor-based controller and amplified into a signal that the experimenter can observe on the level of the behavior (say fluorescence) of the cell. Of course, this requires a ridiculous amount of both fundamental work on quantum computing, and bio-engineering. However, could the future of scalable quantum computers be in the noisy world of biology, instead of the sterility of superconductors, photon benches, or ion-traps?


Adleman, L. M. (1994). Molecular computation of solutions to combinatorial problems. Science, 266(5187), 1021-1023.

Bonnet J, Yin P, Ortiz ME, Subsoontorn P, & Endy D (2013). Amplifying Genetic Logic Gates. Science PMID: 23539178

Childs, A. M. (2009). Universal computation by quantum walk. Physical review letters, 102(18), 180501. [ArXiv pdf]

Childs, A. M., Gosset, D., & Webb, Z. (2013). Universal Computation by Multiparticle Quantum Walk. Science, 339(6121), 791-794. [ArXiv pdf]

Engel GS, Calhoun TR, Read EL, Ahn TK, Mancal T, Cheng YC et al. (2007). Evidence for wavelike energy transfer through quantum coherence in photosynthetic systems. Nature 446 (7137): 782–6.

Păun, G. (2000). Computing with membranes. Journal of Computer and System Sciences, 61(1), 108-143.

Rebentrost, P., Mohseni, M., Kassal, I., Lloyd, S., & Aspuru-Guzik, A. (2009). Environment-assisted quantum transport. New Journal of Physics, 11(3), 033003. [ArXiv pdf]

Tamsir, A., Tabor, J. J., & Voigt, C. A. (2011). Robust multicellular computing using genetically encoded NOR gates and chemical/wires/’. Nature, 469(7329), 212-215.

Programming playground: A whole-cell computational model

Three days ago, Jonathan R. Karr, Jayodita C. Sanghvi and coauthors in Markus W. Covert’s lab published a whole-cell computational model of the life cycle of the human pathogen Mycoplasma genitalium. This is the first model of its kind: they track all biological processes such as DNA replication, RNA transcription and regulation, protein synthesis, metabolism and cell division at the molecular level. To achieve this, the authors integrate 28 different sub-models of the known cellular processes.

Figure 1A from Karr, Sanghvi et al. (2012)

Figure 1A from Karr, Sanghvi et. al (2012): A diagram of the 28 sub-models, colored by category: RNA (green), protein (blue), metabolic (orange), DNA (red). The modules are connected by arrow representing common metabolites (orange), RNA (green), proteins (blue), and DNA (red).

The key technical accomplishment was integrating the 28 modulus into a single model. Each module is based on existing models, but different modules are expressed in different paradigms: ODE, Boolean, probabilistic, and constraint-based. For me, this is the most impressive aspect of the work. Usually, when I look at biology (or psychology), I see a mishmash of models with each expressed in its own language and seemingly incompatible with the others. The authors overcame this by assuming the modulus are independent on short timescales (under 1 second). This allows the software to keep track of 16 global cell variables which are used as inputs for the submodulus that are run to simulate 1 second and their results used to update the global variables and repeat the loop. The whole software is available online and the authors can use the data gathered to produce a video of a single cell’s life cycle:

The authors show that the model has a high level of agreement with existing data. They also use the predictions to run several novel real-biology experiments, and even partially overturn (or complete) a previous experimental observation based on hints from their model. In particular they show that disruption of the IpdA gene — which Glass et al. (2006) suggested as non-essential — has severe (but noncritical) impact on cell growth. I wish I could comment more on the validity of the model as judged by experiments, but molecular biology is magic to me.

The simulation results that were most exciting for me was looking at the effects of single-gene disruptions on phenotype. The bacterium Mycoplasma genitalium is a human urogenital parasite whose genome contains 525 genes (Fraser et al., 1995). It is not an easy model organism to work with, but it has the smallest known genome that can constitute a cell. Part of the team on this project, is from J. Craig Venter Institute and has extensive experience with the organism due to their effort to create the first self-replication synthetic life by implanting artificial DNA into Mycoplasma genitalium. I would not be surprised if this model plays a vital part in the institute’s engineering.

Karr, Sanghvi et al. (2012) ran simulations of each of the 525 possible single-gene disruption strains. They found that 284 genes were essential to sustain growth and division and 117 are non-essential — a 79% agreement with the experimental results of Glass et al. (2006). Of particular interest for me was that in some cases it took more than one generation for specific proteins to fall to lethal levels. As far as I understand this is because when a single-cell divides, daughters get both a copy of the mother DNA and have their initial levels of proteins and RNA set to within statistical fluctuations of those of their mother. Due to my complete lack of basic biological background, this seemed an interesting example of Lamarkian evolution. In particular, it raises questions on how to best combine single-cell learning and evolution. From a naive Bayesian model of learning, it would seem that this would allow cells to pass on their priors — a biological evolution counterpart to Beppu & Griffiths (2009) cultural ratchet.

The detail of the whole-cell model is impressive. I hope that the software becomes a tool for theorists without access to a wet-lab to play around with cells. The approach is an antithesis to the simple and completely unrealistic models I am accustomed to building. For me, it raises many thoughts on how to better think about the distinction between genotype and phenotype that is almost always ignored in evolutionary game theory. For now the whole-cell model is computationally too expensive for me to build evolutionary dynamics from it, but maybe parts of the code can be simplified or ignored or maybe we could use more course-grained models. Either way, I am excited for my new playground!


Beppu, A., & Griffiths, T. (2009). Iterated learning and the cultural ratchet. Proceedings of the 31st Annual Conference of the Cognitive Science Society, 2089-2094.

Fraser, C.M., Gocayne, J.D., White, O., Adams, M.D., Clayton, R.A., Fleischmann, R.D., Bult, C.J., Kerlavage, A.R., Sutton, G., Kelley, J.M., et al. (1995). The minimal gene complement of Mycoplasma genitalium. Science 270, 397–403.

Glass, J.I., Assad-Garcia, N., Alperovich, N., Yooseph, S., Lewis, M.R., Maruf, M., Hutchison, C.A., Smith, H.O., & Venter, J.C. (2006). Essential genes of a minimal bacterium. Proc. Natl. Acad. Sci. USA 103, 425–430.

Karr, J.R., Sanghvi, J.C., Macklin, D.N., Gutschow, M.V., Jacobs, J.M., Bolival, B., Assad-Garcia, N., Glass, J.I., & Covert, M.W. (2012). A whole-cell computational model predicts phenotype from genotype Cell, 150, 389-401 DOI: 10.1016/j.cell.2012.05.044