Developmental modularity may facilitate morphological evolution by allowing phenotypic change of a developing body component without negatively impacting other components. I examined foregut development in Amphissa columbiana, a predatory neogastropod with a highly derived foregut and in Crepidula fornicata, a phytoplankton-feeder with a less derived foregut, for evidence of developmental modules. Histological sections revealed that the post-metamorphic buccal cavity and radula of both species form as a ventral outpocketing (ventral module) from the larval esophagus (dorsal module). However, in Amphissa columbiana the ventral outpocketing is semi-isolated from the larval esophagus and also produces an “anterior esophagus” that is not developmentally homologous to the “anterior esophagus” of herbivorous caenogastropods. Semi-isolation of the ventral and dorsal modules of the developing neogastropod foregut allows precocious development of the post-metamorphic foregut during the larval stage without compromising larval feeding. Therefore, development of diverse variants of the post-metamorphic foregut are freed from larval constraints. / Graduate / 2020-08-03

<div>Multi-model inference covers a wide range of modern statistical applications such as variable selection, model confidence set, model averaging and variable importance.</div><div>The performance of multi-model inference depends on the availability of candidate models, whose quality has been rarely studied in literature. In this dissertation, we study genetic algorithm (GA) in order to obtain high-quality candidate models. Inspired by the process of natural selection, GA performs genetic operations such as selection, crossover and mutation iteratively to update a collection of potential solutions (models) until convergence. The convergence properties are studied based on the Markov chain theory and used to design an adaptive termination criterion that vastly reduces the computational cost. In addition, a new schema theory is established to characterize how the current model set is improved through evolutionary process. Extensive numerical experiments are carried out to verify our theory and demonstrate the empirical power of GA, and new findings are obtained for two real data examples. </div>

Context： Software evolution comes with the increasing growth of software applications both in size and complexity. Unlike the software maintenance, software evolution addresses more on the adaption of the new fast-changing requirements. Then the term of “software evolvability” comes with its importance for evaluating the evolution status of the software. However, it is not clearly identified especially in the context of open source software (OSS). Besides the most studies are about the description of software evolvability as a quality attribute, and very few research have done on the measurement of software evolvability during the software evolution process. Objectives： In this study we perform an in-depth investigation on identification of the OSS evolvability, and figure out the appropriate metrics used for measuring the OSS evolvability. Based on that we finally proposed the open source software evolvability measurement framework (OSEM) which could be used for measuring the software evolvability generally in an OSS context. Methods： At first, we conducted a literature review by combining backward snowballing search with systematic database search. Two research questions which are RQ1 and RQ2 are proposed for helping us to retrieve the key information for building the needed framework. Then we performed a case study on VLC media player (an OSS project) to validate the processes of the proposed framework. Results： Based on literature we could explicitly identify the OSS evolvability, and figure out the differences of software evolvability addressed in OSS context and non OSS context (e.g, the traceability refers to documentation in non OSS context, however in OSS context it refers to the release version of OSS project). Besides we also fulfill the evolvability measuring method by addressing the process of prioritization of evolvability sub-characteristics. In the end we implement the OSEM framework on VLC media player and get the well documented results which are clearly presented and easy to understand. Such results could be taken by the VLC developers as an input for the design and development of the VLC. Conclusions: We conclude that the open source software measurement framework (OSEM) is applicable, based on the time we spent on the case of VLC media player it is quite fast and efficient to use such framework. The results from the conduction of this framework are documented well and very clear for OSS users/developers to follow.

The aim of this thesis is to investigate aspects of molecular evolution and enzyme engineering using the experimental evolution of Tobacco Etch Virus cysteine protease (TEV) as a model. I map key features of the local fitness landscape and characterise how they affect details of enzyme evolution. In order to investigate the evolution of core active site machinery, I mutated the nucleophile of TEV to serine. The differing chemical properties of oxygen and sulphur force the enzyme into a fitness valley with a >104-fold activity reduction. Nevertheless, directed evolution was able to recover function, resulting in an enzyme able to utilise either nucleophile. High-throughput screening and sequencing revealed how the array of possible beneficial mutations changes as the enzyme evolves. Potential adaptive mutations are abundant at each step along the evolutionary trajectory, enriched around the active site periphery. It is currently unclear how seemingly neutral mutations affect further adaptive evolution. I used high-throughput directed evolution to accumulate neutral variation in large, evolving enzyme populations and deep sequencing to reconstruct the complex evolutionary dynamics within the lineages. Specifically I was able to observe the emergence of robust enzymes with improved mutation tolerance whose descendants overtake later populations. Lastly, I investigate how evolvability towards new substrate specificities changed along these neutral lineages, dissecting the different determinants of immediate and long-term evolvability. Results demonstrate the utility of evolutionary understanding to protease engineering. Together, these experiments forward our understanding of the molecular details of both fundamental evolution and enzyme engineering.

Nowadays, while software industries are aiming to develop their software continuously, their delivery is hindered by technical debt.  Preventing technical debt would be valuable if it is considered in architectural decisions. On the other side, since microservices architecture is adaptable to build cloud applications and has other advantages, it has become a trend in the software industries. Due to the popularity of microservices and the importance of technical debt in the software industry, this research aims to find the influence of architectural decisions on technical debt in microservices applications. In this research, we explore architectural decisions in microservice applications and their qualities that impact technical debt.   We calculated the repetitiveness of selected microservices architectural decisions and the extra effort that they need to meet qualities to prevent technical debt. Spearman correlation coefficient used to calculate the relation between extra effort on the qualities of architectural decisions in microservice applications that affect technical debt. Furthermore, we calculated the correlation between the repetitiveness of selected architectural decisions and the effort for their qualities to find the effect of repetitiveness on qualities that reduce technical debt.   Our result shows that every architectural decision that we have explored for microservice applications needs some extra effort to increase the quality that can prevent technical debt. Correlation between qualities and repetitiveness of architectural decisions shows that weak correlation, which proves that increasing or decreasing of repetitiveness would not change the demand for extra effort to prevent technical debt.

System evolvability is vital to the longevity of large-scale complex engineered systems. The need for evolvability in complex systems is a result of their long service lives, rapid advances to their integrated technologies, unforeseen operating conditions, and emerging system requirements. In recent years, quantifiable metrics have been introduced for measuring the evolvability of complex systems based on the amount of excess capability in the system. These metrics have opened opportunities for optimization of systems with evolvability as an objective. However, there are several aspects of such an optimization that require further consideration. For example, there is a trade-off between the cost of excess capability initially built into complex systems and the benefit that is added to the system for future evolution. This trade-off must be represented in the optimization problem formulation. Additionally, uncertainty in future requirements and parameters of complex systems can result in an inaccurate representation of the design space. This thesis addresses these considerations through multi-objective optimization and uncertainty analysis. The resulting analysis gives insight into the effects of designing for evolvability. We show that there is a limit to the value added by increasing evolvability. We also show that accounting for uncertainty changes the optimal amount of evolvability that should be designed into a system. The developed theories and methods are demonstrated on the design of a military ground vehicle.

No description available.

La vie sur Terre s’est adaptée à un environnement où il y a un faible et persistent bruit de fond radiatif qui interagit avec les cellules. Loin des effets clairement nocifs des radiations à haute dose, il est difficile d’évaluer et de comprendre les impacts des faibles doses de la radioactivité naturelle sur les systèmes vivants. Nous avons tenté d’étudier si le bruit de fond radiatif est un facteur important dans l’évolution, en menant des expériences évolutives identiques avec Escherichia coli au Laboratoire de Physique Corpusculaire de Clermont-Ferrand, et au Laboratoire Souterrain de Modane. Malgré une différence d’un facteur 7,3 entre les taux d’interaction des rayonnements ionisants avec les cellules dans les deux laboratoires, aucune différence significative n’a pu être trouvée dans le fitness compétitif des populations cellulaires évoluées dans chaque laboratoire. Par simulation, nous avons montré que le taux d’interaction entre le bruit de fond radiatif et E. coli est cent fois plus faible que le taux de mutations d’origine endémique, ce qui renforce l’hypothèse que les radiations naturelles ont peu d’effet sur l’évolution. Dans le cadre du projet Geant4-DNA, nous avons développé une application complète de simulation mécanistique des dommages radio-induits à l’ADN, afin d’explorer davantage cette hypothèse. Avec cette application, on a irradié un modèle du génome d’E. coli, montrant que pour l’irradiation par des électrons d’énergies > 10 keV, le rendement des cassures double brin est de 0,006 – 0,010 CDB Gy-1 Mbp-1, selon le modèle de piégeage des radicaux chimiques. Ce résultat est en accord avec des données expérimentales, et souligne plus encore que les radiations ionisantes d’origine naturelle n’ont qu’une contribution mineure aux mutations responsables de l’évolution. / All life on earth has adapted to an environment where there is a small, persistent, radiation background interacting with cells. Unlike evaluating the clearly harmful effects of high radiation doses, understanding the effects of this low persistent radiation dose on living systems is incredibly difficult. We have attempted to study whether background radiation is an important factor in evolution by conducting identical evolution experiments with Escherichia coli in the Clermont-Ferrand Particle Physics Laboratory and the Modane Underground Laboratory. Despite a 7.3 fold difference in the rate of interactions between the radiation background and cells between the two environments, no significant difference was found in the competitive fitness of the cell populations grown at each location. Using simulations, we showed that the rate at which ionising radiation interacts with cells is one hundred times less frequent than E. coli’s mutation rate in our experimental conditions, supporting the contention that natural radiation has no strong evolutionary effect. To further support this conclusion, we developed a mechanistic simulation for DNA damage as part of the Geant4-DNA project. Using this application, we irradiated a model of an E. coli genome, showing that for electron irradiation > 10 keV, the double strand break yield can be reasonably estimated to be between 0.006 – 0.010 DSB Gy-1 Mbp-1, depending upon the modelling of radical scavenging. This is in agreement with experimental data, further highlighting the small role natural ionising radation plays as a cause of mutations.

An important factor in system longevity is service-phase evolvability, which is defined as the ability of a system to physically transform from one configuration to a more desirable configuration while in service. These transformations may or may not be known during the design process, and may or may not be reversible. A study of 210 engineered systems was performed and found that system excess and modularity allow a system to evolve while in service. Building on these observations, this thesis introduces mathematical relationships that map a system's excess and modularity to that system's ability to evolve. These relationships are derived from elastic potential energy theories. The use of the evolvability measure, and other related measures presented herein, are illustrated with simple numerical examples and applied to the design of US Navy nuclear aircraft carriers. Using these relationships, it is shown that the Navy's new Ford-class aircraft carrier is the most evolvable carrier designed to date. Though the evolvability relationships introduced here are generically derived based on excess and modularity, the aircraft carrier example presented considers only the system excess.

The frozen plasticity theory is a punctuationalist theory of adaptive evolution. It states that long periods of stasis, during which populations respond to selection pressures only by elastic change in the frequency of already present alleles, alternate in the evolution of sexual species with short periods of plastic evolution, during which alleles can get fixed or eliminated by directed selection. Asexual species are not expected to maintain such high genetic polymorphism in the long term. They should, however, be able to plastically respond to selection pressures throughout their whole existence. This difference between the evolutionary dynamics of sexual and asexual clades has a number of ecological and macroevolutionary implications. Concerning ecology, we could expect different environmental preferences of sexual and asexual species. Accordingly, in our first work that was based on a comparative study, we statistically significantly supported the hypothesis that (ancient) asexual groups of (eukaryotes) inhabit more stable and homogeneous habitats than their related sexual controls. Focusing on actually experienced, i.e. subjective, heterogeneity of the environment turned out to be the crucial factor of this type of research. From the viewpoint of macroevolutionary implications of the frozen...