Search for: All records

Spatial structure is hypothesized to be an important factor in the origin of life, wherein encapsulated chemical reaction networks came together to form systems capable adaptive complexification via Darwinian evolution. In this work, we use a computational model to investigate how different patterns of environmental connectivity influence the emergence of adaptive processes in simulated systems of self-amplifying networks of interacting chemical reactions (autocatalytic cycles, “ACs”). Specifically, we measured the propensity for adaptive dynamics to emerge in communities with nine distinct patterns of inter-AC interactions, across ten different patterns of environmental connectivity. We found that the pattern of connectivity can dramatically influence the emergence of adaptive processes; however, the effect of any particular spatial pattern varied across systems of ACs. Relative to a well-mixed (fully connected) environment, each spatial structure that we investigated amplified adaptive processes for at least one system of ACs and suppressed adaptive processes for at least one other system. Our findings suggest that there may be no single environment that universally promotes the emergence of adaptive processes in a system of interacting components (e.g., ACs). Instead, the ideal environment for amplifying (or suppressing) adaptive dynamics will depend on the particularities of the system. 

Abstract Genetic Programming (GP) often uses large training sets and requires all individuals to be evaluated on all training cases during selection. Random down-sampled lexicase selection evaluates individuals on only a random subset of the training cases, allowing for more individuals to be explored with the same number of program executions. However, sampling randomly can exclude important cases from the down-sample for a number of generations, while cases that measure the same behavior (synonymous cases) may be overused. In this work, we introduce Informed Down-Sampled Lexicase Selection. This method leverages population statistics to build down-samples that contain more distinct and therefore informative training cases. Through an empirical investigation across two different GP systems (PushGP and Grammar-Guided GP), we find that informed down-sampling significantly outperforms random down-sampling on a set of contemporary program synthesis benchmark problems. Through an analysis of the created down-samples, we find that important training cases are included in the down-sample consistently across independent evolutionary runs and systems. We hypothesize that this improvement can be attributed to the ability of Informed Down-Sampled Lexicase Selection to maintain more specialist individuals over the course of evolution, while still benefiting from reduced per-evaluation costs. 

Humans have long known how to co-opt evolutionary processes for their own benefit. Carefully choosing which individuals to breed so that beneficial traits would take hold, they have domesticated dogs, wheat, cows and many other species to fulfil their needs. Biologists have recently refined these ‘artificial selection’ approaches to focus on microorganisms. The hope is to obtain microbes equipped with desirable features, such as the ability to degrade plastic or to produce valuable molecules. However, existing ways of using artificial selection on microbes are limited and sometimes not effective. Computer scientists have also harnessed evolutionary principles for their own purposes, developing highly effective artificial selection protocols that are used to find solutions to challenging computational problems. Yet because of limited communication between the two fields, sophisticated selection protocols honed over decades in evolutionary computing have yet to be evaluated for use in biological populations. In their work, Lalejini et al. compared popular artificial selection protocols developed for either evolutionary computing or work with microorganisms. Two computing selection methods showed promise for improving directed evolution in the laboratory. Crucially, these selection protocols differed from conventionally used methods by selecting for both diversity and performance, rather than performance alone. These promising approaches are now being tested in the laboratory, with potentially far-reaching benefits for medical, biotech, and agricultural applications. While evolutionary computing owes its origins to our understanding of biological processes, it has much to offer in return to help us harness those same mechanisms. The results by Lalejini et al. help to bridge the gap between computational and biological communities who could both benefit from increased collaboration. 

Short Abstract for evolutionary computing community on selection schemes working for directed evolution experiments in microbes. 

We introduce and experimentally demonstrate the utility of tag-based genetic regulation, a new genetic programming (GP) technique that allows programs to dynamically adjust which code modules to express.Tags are evolvable labels that provide a flexible mechanism for referencing code modules. Tag-based genetic regulation extends existing tag-based naming schemes to allow programs to “promote” and “repress” code modules in order to alter expression patterns. This extension allows evolution to structure a program as a gene regulatory network where modules are regulated based on instruction executions. We demonstrate the functionality of tag-based regulation on a range of program synthesis problems. We find that tag-based regulation improves problem-solving performance on context-dependent problems; that is, problems where programs must adjust how they respond to current inputs based on prior inputs. Indeed, the system could not evolve solutions to some context-dependent problems until regulation was added. Our implementation of tag-based genetic regulation is not universally beneficial, however. We identify scenarios where the correct response to a particular input never changes, rendering tag-based regulation an unneeded functionality that can sometimes impede adaptive evolution. Tag-based genetic regulation broadens our repertoire of techniques for evolving more dynamic genetic programs and can easily be incorporated into existing tag-enabled GP systems. 

Symbiosis, the living together of unlike organisms as symbionts, is ubiquitous in the natural world. Symbioses occur within and across all scales of life, from microbial to macro-faunal systems. Further, the interactions between symbionts are multimodal in both strength and type, can span from parasitic to mutualistic within one partnership, and persist over generations. Studying the ecological and evolutionary dynamics of symbiosis in natural or laboratory systems poses a wide range of challenges, including the long time scales at which symbioses evolve de novo , the limited capacity to experimentally control symbiotic interactions, the weak resolution at which we can quantify interactions, and the idiosyncrasies of current model systems. These issues are especially challenging when seeking to understand the ecological effects and evolutionary pressures on and of a symbiosis, such as how a symbiosis may shift between parasitic and mutualistic modes and how that shift impacts the dynamics of the partner population. In digital evolution, populations of computational organisms compete, mutate, and evolve in a virtual environment. Digital evolution features perfect data tracking and allows for experimental manipulations that are impractical or impossible in natural systems. Furthermore, modern computational power allows experimenters to observe thousands of generations of evolution in minutes (as opposed to several months or years), which greatly expands the range of possible studies. As such, digital evolution is poised to become a keystone technique in our methodological repertoire for studying the ecological and evolutionary dynamics of symbioses. Here, we review how digital evolution has been used to study symbiosis, and we propose a series of open questions that digital evolution is well-positioned to answer. 

Fluctuating environmental conditions are ubiquitous in natural systems, and populations have evolved various strategies to cope with such fluctuations. The particular mechanisms that evolve profoundly influence subsequent evolutionary dynamics. One such mechanism is phenotypic plasticity, which is the ability of a single genotype to produce alternate phenotypes in an environmentally dependent context. Here, we use digital organisms (self-replicating computer programs) to investigate how adaptive phenotypic plasticity alters evolutionary dynamics and influences evolutionary outcomes in cyclically changing environments. Specifically, we examined the evolutionary histories of both plastic populations and non-plastic populations to ask: (1) Does adaptive plasticity promote or constrain evolutionary change? (2) Are plastic populations better able to evolve and then maintain novel traits? And (3), how does adaptive plasticity affect the potential for maladaptive alleles to accumulate in evolving genomes? We find that populations with adaptive phenotypic plasticity undergo less evolutionary change than non-plastic populations, which must rely on genetic variation from de novo mutations to continuously readapt to environmental fluctuations. Indeed, the non-plastic populations undergo more frequent selective sweeps and accumulate many more genetic changes. We find that the repeated selective sweeps in non-plastic populations drive the loss of beneficial traits and accumulation of maladaptive alleles, whereas phenotypic plasticity can stabilize populations against environmental fluctuations. This stabilization allows plastic populations to more easily retain novel adaptive traits than their non-plastic counterparts. In general, the evolution of adaptive phenotypic plasticity shifted evolutionary dynamics to be more similar to that of populations evolving in a static environment than to non-plastic populations evolving in an identical fluctuating environment. All natural environments subject populations to some form of change; our findings suggest that the stabilizing effect of phenotypic plasticity plays an important role in subsequent adaptive evolution.