Philosophiae Doctor - PhD / In this thesis, we present a multi-wavelength analysis of star-forming galaxies to shed new light on the evolution of the far-IR-radio relations in intermediate redshift (0.3 < z < 0.6) galaxy clusters and galaxy groups. The far-infrared (far-IR) emission from galaxies is dominated by thermal dust emission. The radio emission at 1.4 GHz is predominantly produced by non-thermal synchrotron radiation. The underlying mechanisms, which drive the far-IR-radio correlation, are believed to arise from massive star formation. A number of studies have investigated the relationship as a function of redshift in the field and have found no evolution out to at least z _ 2, however few works have been done in galaxy clusters. In nearby clusters, the median logarithmic ratio of the far-IR to radio luminosity is qFIR = 2.07_0.74, which is lower than the value found in the field, and there is an indication of an enhancement of radio emission relative to the far-IR emission. Understanding the properties of the far-IR-radio correlation in a sample of distant and massive cluster and groups plays an important role in understanding the physical processes in these systems. We have derived total infrared luminosities for a sample of cluster, group, and field galaxies through an empirical relation based on Spitzer MIPS 24 _m photometry. The radio flux densities were measured from deep Very Large Array 1.4 GHz radio continuum observations. We have studied the properties of the far-IR-radio correlation of galaxies at intermediate redshift clusters by comparing the relationship of these galaxies to that of low redshift clusters. We have also examined the properties of the galaxies showing radio excess to determine the extent that galaxy type or environment may explain the radio excess in galaxy clusters. We find that the ratio of far-IR to radio luminosity for galaxies in an intermediate redshift cluster to be qFIR = 1.72_0.63. This value is comparable to that measured in low redshift clusters. A higher fraction of galaxies in clusters show an excess in their radio fluxes when compared to low redshift clusters, and corroborates previous evidence of a cluster enhancement of radio excess sources at this earlier epoch as well. We have also investigated the properties of the far-IR-radio correlation for a sample of galaxy groups in the COSMOS field. We find a lower percentage of radio-excess sources in groups as compared to clusters. This provides preliminary evidence that the number of radioexcess sources may depend on galaxy environment. We also find that a larger fraction of radio-excess sources in clusters are red sequence galaxies.

Millisecond pulsars (MSP) are old neutron stars that have been spun up to high spin frequencies(as fast as 716 Hz) through the accretion of matter from a companion star. The extreme stellar densities in the core of globular clusters creates numerous accreting neutron star systems through exchange interactions; this leads to the formation of MSPs in larger numbers than in the galactic disk. Over the course of this project, we have collected over 17 TB of data on the 3 globular clusters M28, NGC6440 and NGC6441 plus 2 observations on NGC6522 and NGC6624 as part of the recently begun S-band survey using the Green Bank telescope. I have analyzed and conducted acceleration searches on 70% of the data and discovered 7 of the 23 new millisecond pulsars reported in this work. One year of timing observations of the pulsars in M28 and NGC6440 has led to the phase connected solution for 12 of the 15 new pulsars in those two clusters, 7 of which are in binaries. We have measured the rate of advance of periastron for two highly eccentric binaries and assuming this is purely due to general relativity, this leads to total system masses of (1.616 - 0.014)M and (2.2 - 0.8)M for M28C and NGC6440B respectively. The small mass function combined with this information imply that the most likely neutron star mass of NGC6440B is either very large or else there could be significant contribution to the advance of periastron from a nonzero quadrupole moment due to tidal interaction with the companion. Measurements of the period derivatives for many of the pulsars show that they are dominated by the dynamical effect of the gravitational field of the clusters. Finally, we have discovered the potential presence of a Mars-mass planet orbiting the pulsar NGC6440C with a period of 21 days. A dedicated timing campaign will be necessary to confirm the presence of such an object. / Science, Faculty of / Physics and Astronomy, Department of / Graduate

Clusters of atoms have remarkable optical properties that were exploited since the antiquity. It was only during the late 20th century though that their production was better controlled and opened the door to a better understanding of matter. Lasers are the tool of choice to study these nanoscopic objects so scientists have been blowing clusters with high intensities and short duration laser pulses to gain insights on the dynamics at the nanoscale. Clusters of atoms are an excellent first step in the study of bio-molecules imaging. New advancements in laser technology in the shape of Free Electron Lasers (FEL) made shorter and shorter wavelengths accessible from the infrared (IR) to the vacuum and extreme ultra-violet (VUV and XUV) to even X-rays. Experiments in these short wavelengths regimes revealed surprisingly high energy absorption that are yet to be fully explained. This thesis tries to increase the global knowledge of clusters of rare-gas atoms interacting with short duration and high intensity lasers in the VUV and XUV regime. Theoretical and numerical tools were developed and a novel model of energy transfer based on excited states will be presented. The first part describes the current knowledge of laser-cluster interaction in the short wavelength regime followed by the description of the new model. In the second part of the thesis the different tools and implementations used throughout this work are presented. Third, a series of journal articles (of which four are published and one to be submitted) are included where our models and tools were successfully used to explain experimental results.

Les amas de galaxies sont l'une des sondes cosmologiques permettant de contraindre les modèles d'univers, via leur fonction de masse et leur vitesse de croissance, elles-mêmes mesurées par l'état dynamique des amas. Les grands relevés présents et futurs permettent d'avoir accès à une information plus nombreuse et plus complète sur les amas de galaxies et donc d'utiliser de nouvelles méthodes de détermination de leur état dynamique. Dans cette thèse, une nouvelle méthode de caractérisation morphologique 2+1D des amas a été développée afin d'établir une nouvelle classification des amas. Il s'agit d'une méthode optique basée sur la position et la vitesse radiale des galaxies. Les structures dans la zone d'influence des amas sont détectées et caractérisées en projection et dans l'espace des vitesses radiales à l'aide d'une analyse en ondelettes. À partir du nombre de structures, les amas sont classés en amas unimodal, bimodal ou multimodal. L'ellipticité de leur distribution projetée et la gaussianité de la distribution des vitesses radiales sont également utilisées pour raffiner la classification. La méthode de caractérisation et de classification morphologique a été appliquée à un sous-échantillon de 403 amas issus du catalogue C4 en utilisant les données du SDSS. Il en est ressorti que 25% des amas sont unimodaux, 33% sont bimodaux et 42% sont multimodaux. Une analyse de la stabilité de la classification a également été réalisée ainsi qu'une comparaison avec les résultats de la littérature, que ce soit d'un point de vue statistique ou au niveau des amas individuels. / Clusters of galaxies are one of the main cosmological probes used to constrain the cosmological parameters, through their mass function and their growth rate. The measure of these two quantities require the determination of the dynamical state of clusters. The present and future large and deep sky surveys give access to a more complete information on clusters and legitimate the development of new methods of determination of their dynamical state. In this thesis, a new method of characterization of the cluster morphology has been developed. It is a 2+1D method using galaxies and it enables to develop a new morphological classification of clusters. Structures around clusters are detected and characterized in projection and along the line of sight using a wavelet analysis. The new classification consists in counting the number of structures in the vicinity of clusters, three clusters classes were defined : unimodal, bimodal and multimodal. The ellipticity and the Gaussianity of the distribution of radial velocities are also used to refine the classification. The method was applied to a subsample of 403 clusters from the C4cluster catalogue using data from the SDSS. The results are : 25% of the clusters are unimodals, 33% are bimodals and 42% are multimodals. The stability of the classification with respect to the different parameters used was also performed as well as a comparison with the results from other studies in the literature.

Businesses cluster to achieve agglomeration benefits. However, research in developing countries suggests that the economic environment limits small business’ propensity to benefit from agglomerations. The study examines the location, networking patterns, formal structures and owner characteristics of 1256 micro businesses from ten industries and thirteen sample areas in Toluca, Mexico. First, the thesis analyses whether clustering has a positive impact on the success rates of the surveyed enterprises, e.g. higher sales per employee. On an industry scale only Retail benefits from agglomerations economies. However, results of the neighborhood data show that specific areas benefit from urbanization economies. Overall, the study finds that businesses located within agglomerations, have higher levels of formalization, networking and professional training, hence constituting a more sophisticated base for economic development. Conclusions can be drawn for development policies and programs, arguing for a more differentiated approach of small business development depending on business location and cluster characteristics.

Despite their homology and analogous function, the Hox gene clusters of vertebrates and invertebrates are subject to different constraints on their structural organization. This is demonstrated by a drastically different distribution of repetitive DNA elements in the Hox cluster regions. While gnathostomes have a strong tendency to exclude repetitive DNA elements from the inside of their Hox clusters, no such trend can be detected in the Hox gene clusters of protostomes. Repeats “invade” the gnathostome Hox clusters from the 5′ and 3′ ends while the core of the clusters remains virtually free of repetitive DNA. This invasion appears to be correlated with relaxed constraints associated with gene loss after cluster duplications.

Identifying biosynthetic gene clusters from genomic data is challenging, with many in-silico tools suffering from a high rediscovery rate due to their dependence on rule-based algorithms. Next generation sequencing has provided an abundance of genomic information, and it has been hypothesized that there are many undiscovered biosynthetic gene clusters within this dataset. Here, we aim to develop a machine learning tool, ML-Miner, that infers patterns that describe a biosynthetic gene cluster in an unbiased manner and, as such, enables the identification of new biosynthetic gene clusters from genomic data. To solve this challenging problem, we define a simpler one to predict the class of a known BGC. Specifically, ML-Miner receives as input the concatenation of sequences that are known or believed to be part of a biosynthetic gene cluster. Its task is to identify which class it belongs, i.e. NPRS, PKS terpene and RiPPs. ML-Miner is a machine learning tool that uses Natural Language Processing, dimensionality reduction, and supervised learning to identify novel biosynthetic gene clusters. BioVec is a biological word embedding that we use to transform protein sequences from the highly curated MIBiG database of characterized biosynthetic gene clusters into their respective continuous distributed vector representations. Because the resulting protein vectors are of high dimensionality, a supervised Uniform Manifold and Approximation algorithm was employed to transform the high dimensional vectors into a robust lower-dimensional representation, as evaluated by Silhouette analysis, Hopkins’ statistic, and trustworthiness analysis. The density-Based Spatial Clustering of Applications and Noise algorithm showed that the clusters identified from the low dimensional datasets mapped to biosynthetic gene cluster types, defined with high accuracy in the MIBiG database. A random forest classifier was then trained and evaluated using the low dimensional vectors. It was shown to classify each biosynthetic gene cluster from the MIBiG database with excellent performance metrics. Finally, the model's ability to generalize was evaluated using biosynthetic gene clusters from the antiSMASH dataset, an uncurated database containing uncharacterized biosynthetic gene clusters. The performance metrics were high, with a balanced accuracy of ~85%. After a hyperparameter search, the balanced accuracy rose to ~90%. This suggests that ML-Miner is a robust machine learning pipeline that can be used to identify novel biosynthetic gene clusters. Future development of a confidence score for classification and a workflow for processing bacterial genomes into gene clusters will significantly improve the utility of this tool.

No description available.

Most star formation in galaxies takes place in embedded clusters, within Giant Molecular Clouds (GMCs). Stars also generally form as part of binary star systems, with almost all massive stars having at least one close companion. Binaries shape the physical properties of older star clusters by setting their central density and ejecting low-mass stars, but also play a role during cluster formation by modifying the mechanical and radiative feedback from massive stars and shedding enriched material in the cluster’s gas reservoir. Conversely, dynamical interactions between stars in dense stellar environments are known to form, modify, and destroy binary systems. In consequence, the populations of binaries observed in the Galactic field and in old stellar clusters are understood to be shaped by a combination of the physics of star formation and subsequent dynamical interactions in embedded clusters, although the relative importance of these processes remains unknown. In this thesis, we implement a prescription for an initial population of binaries in the coupled N-body and radiation hydrodynamics star cluster formation code Torch, and investigate how this initial population is modified in the earliest stages of cluster formation, while gas and stars coexist. As an ansatz for the initial population of binaries, we use the properties of main-sequence binaries in the Galactic field. We first perform a suite of simulations initialized from a 10^4 M⦿ cloud, in which the simulations only differ by their stellar content (i.e. presence or absence of an initial population of binaries, and stochasticity of star formation). We compare the populations of binaries identified 1.2–2 Myr after the onset of star formation and find that an initial population of binaries is needed at all masses to reproduce the multiplicity fraction observed in main-sequence stars. We also show that this initial population is modified in a systematic manner before the effects of feedback from massive stars shape the gas. We further find evidence of both preferential formation and preferential destruction of binaries via dynamical interactions. The net effect of these interactions shifts the distributions of primary masses and semi-major axes to lower values, and the distributions of mass ratios and eccentricities to larger values. In a second time, we perform simulations with different virial parameters and initial turbulent velocity patterns, and find that the trends previously identified are robust to those changes in our initial conditions. We however find that both the virial parameter and the initial turbulent velocity pattern have a strong influence on the star formation rate, and therefore on the rapidity with which the distributions are modified. We conclude that dynamical interactions in embedded clusters are important for shaping the populations of binaries observed in the MilkyWay, thus opening the floor to future investigations of the impact of binaries on star cluster formation. / Thesis / Master of Science (MSc)

Star formation is an inherently multi-scale process, connecting scales from the kiloparsecs of the galactic disk to the single AU scale of a protostar. In the middle of these scales are star clusters and molecular clouds, the structures in which most stars form. The clouds and clusters are connected via the interstellar medium, the gas and dust making up the matter between stars. In the cold phases of this medium rests the first steps of star formation, the formation of molecular gas and networks of filaments. This cold, neutral medium (CNM) hosts a handful of physical mechanisms, all contributing to the structures that feeds star formation. In this thesis work, we present a suite of simulations using the magneto-hydrodynamical code Ramses to investigate the role of turbulence, magnetic fields and cooling on the formation of filaments and clusters in the CNM. Through 9 different models we find that velocity dispersions in the CNM play a significant role in the formation of structure, requiring a balance between turbulence, self gravity and cooling to form filaments. We find magnetic fields, initialized at strengths of 7 muG, affect the formation of filaments, creating higher percentages of star-forming dense gas and lower percentages of molecular gas. Both magnetic fields and velocity dispersion in the gas affect the formation rate of clusters early in the simulation. Our 8 km/s simulations present a good initial condition for star formation that can include multiple scales of the process and recreate accurate clouds and filamentary structure. / Thesis / Master of Science (MSc)