Modern scientific methods heavily rely on large-scale 3D simulations. However, current data production speeds are much higher than storage speeds (5 orders of magnitude on typical supercomputers). This imbalance constitutes a major bottleneck in the scientific computing pipeline, such that most of the data generated by a simulation is not saved to disk, and thus remains unvisualized, unexplored and unanalyzed. TORI addresses this data bottleneck by introducing the next generation data reduction tools for large-scale scientific 3D data. TORI’s angle of attack is based on original and important advances in Topological Data Analysis (TDA), a class of techniques popularized in scientific visualization. TORI addresses data reduction at two levels: (i) at the data level, by deriving an analysis framework for ensembles of topological objects that is inspired by optimal transport, and (ii) at the computation level, by entirely revisiting TDA to adapt it to the context of high-performance in-situ data analysis. To identify informative datasets (i), TORI will introduce efficient methods for distance computations, barycenter evaluations and trajectory analysis. To perform this analysis on-the-fly (ii), TORI will revisit TDA with task parallel algorithms, coarse-to-fine computations and TDA-aware lossy compressors. TORI will be implemented in open-source in the Topology ToolKit, a leading TDA package, and interfaced with standard scientific computing packages (VTK, ParaView). It will be integrated in simulation codes with Catalyst and evaluated on real-life use cases in climatology, geophysics and astrophysics. TORI will have a far reaching impact on all fields of science using large-scale 3D simulations. By bringing together optimal transport and TDA in an innovative coarse-to-fine model, TORI will establish TDA as a standard tool for the analysis of large-scale ensemble datasets and it will initiate a new line of research in high-performance in-situ data analysis.

Scattering of light in complex environments has long been considered a nuisance and an inescapable limitation to imaging and sensing alike, ranging from astronomical observation, biomedical imaging, spectroscopy, etc. In the last decade, wavefront shaping techniques have revolutionized this view, by allowing light focusing and imaging even deep in the multiple scattering regime. This principle is embodied in the possibility—that I pioneered—to access the transmission matrix of a complex medium. In SMARTIES, I will go one major conceptual step further, by exploiting directly the inherent property of a complex medium to mix perfectly and deterministically the information carried by the light. This mixing is actually a processing step. Along this general idea, SMARTIES will explore two synergistic directions: —Classical and quantum optical computing: Thanks to the highly multimode nature and the strong mixing properties of complex material, I will aim at demonstrating high performance classical computing tasks in the context of randomized algorithms. As a platform for quantum information processing, this will be relevant for high dimension quantum computing algorithms, and quantum machine learning. —Generalized imaging and sensing: Rather than tediously focusing and imaging through a scattering material, computational approaches can significantly improve and simplify the imaging process. I also aim to show that the relevant information can be directly and optimally extracted from the scattered light without imaging, using machine-learning algorithms. From a methodological standpoint, SMARTIES will require bridging knowledge from mesoscopic physics, light-matter interaction, linear and non-linear optics, with algorithms and signal processing concepts. It will deliver a whole new class of optical methods and devices, based on disorder. Its applications range from big data analysis, quantum technologies, to sensors and deep imaging for biology and neuroscience.

Fisheries are operating worldwide over nation’s Economic Exclusive Zones (EEZ) as well as over international waters. Information on the location of fishing is rarely known, especially in international waters, yet it is critical since in many oceanic sectors non declared and illegal fisheries are affecting negatively ecosystems through over exploitation and by catch of non-target species Knowledge about the distribution of fishing boats is fundamental for the regulation of fishing activities as well for the conservation of the oceans. I propose a New Concept of ocean surveillance based on new bio-logging technologies fitted on large foraging marine predators. The OCEAN SENTINEL Proof of this Concepts will (1) develop a logger called CENTURION that couple a XGPS platform detecting and locating radar emissions, with a satellite transmission system (Argos) that would send instantaneously the location of vessels to a receiving site, (2) deploy the logger on wide ranging animals used as platforms and (3) make available immediately the information obtained from the CENTURION logger through a website. The OCEAN SENTINEL programme will generate important social benefits by providing information to a wide range of beneficiaries, from governments or regional authorities managing EEZ and natural resources, regional or national fishing authorities, researchers and non-governmental organisation in conservation. The concept will be tested in the Southern Indian Ocean from Crozet and Kerguelen Islands where valuable and extensive fisheries operate in EEZs and over oceanic waters. This concept has the potential to be used in other areas where information on fisheries location is needed. The project will also lead to further discoveries on the relationship between seabirds and fisheries as well on the extent of fisheries in zones where surveillance by conventional methods is not possible.

The exuberant proliferation of herbivorous insects is often attributed to their association with plants, making their interactions of particular importance to understanding what is driving their vast diversity. Early biologists exploring the underlying factors proposed the hypothesis of coevolution (and the escape and radiate model). Despite general support for this hypothesis, the macroevolutionary and genomic consequences of the origins and evolutionary dynamics of host-plant shifts remain elusive. Recent results illustrate the need for a multidisciplinary approach to assessing the role of host plants in shaping insect diversity at macroevolutionary scales. Using the swallowtail butterflies (Papilionidae) and their host plants, this project will develop a macroevolutionary and genomic framework to studying the origin and evolution of an arms race through time and space. We will build a complete species-level phylogeny for Papilionidae relying on whole-genome sequencing for all species. This time-calibrated phylogeny will be combined with species traits to estimate ancestral host-plant preferences and subsequent host-plant shifts. We will reconstruct dated phylogenies of the main host-plant families to estimate whether the butterflies and their host plants diversified concurrently through time and space. Diversification rates will be estimated for shifting/non-shifting and prey/non-prey clades. A matching genomic survey will to look for genes under positive selection by comparing sets of phylogenetic branches that experienced a host-plant shift versus branches without such a shift. Transcriptomes will be characterized for caterpillars and their plants to identify and pinpoint the genes involved in the arms race, as well as to compare them across the swallowtail tree of life. With this ambitious research proposal, we aim to provide answers to longstanding and fundamental evolutionary questions on the mechanisms behind ecological interactions over long timescales.

In the last decades, the progress in medicine have contributed to extend human lifespan considerably. Unfortunately, this aging of populations gives rise to the increased appearance of degenerative diseases and cancers, and therefore represents a critical problem for public health. During embryonic development, cellular proliferation is an essential process that is required to generate all the tissues and organs that contribute to the adult organism. However, alterations in the molecular mechanisms regulating this fundamental process can drive developmental defects and tumors. Cellular fail-safe systems evolved to prevent these events to happen. One of them is the natural shortening of chromosome ends, termed the telomeres. This “molecular clock” prevents cells from proliferating if their telomeres become critically short, which could lead to chromosomal alterations. This mechanism is responsible for cellular aging in Human. Unfortunately, cancer cells can bypass this process through the activation of telomere maintenance mechanism allowing for virtually unlimited proliferation. Interestingly, such maintenance mechanisms naturally occur during early embryonic development. The proposed research is aimed at understanding how telomeres are regulated in this system and under a physiological context. My expertise in mouse embryonic work associated with the competences in telomere biology of the host laboratory of Dr. Jerome Déjardin will likely be a suitable combination to answer these questions. On the long term, these researches could lead to the discovery of key factors regulating telomere maintenance. Finally, as the suppression of telomere length maintenance inhibits cell growth, the present proposal could contribute to the development of new treatments against cancer or precocious aging.