Talks

We explore emergent dynamics in a continuum model of rigid filament assemblies inspired by the actin-based lamellipodium, a key structure in cellular motility. The model incorporates adhesion, inter-filament pressure, and center-of-mass tension within a quasi-static force balance framework, yielding a variational structure that governs filament strip configurations. By varying stiffness parameters, we uncover pitchfork bifurcations that signal transitions from symmetric, stationary states to asymmetric, motile regimes. This symmetry-breaking behavior captures a fundamental mechanism of spontaneous translocation driven by mechanical cues. Our findings highlight how bifurcation analysis can elucidate critical thresholds in biophysical systems and provide insight into the self-organized motion of cytoskeletal structures.

TBA

The adoption of electric vehicles (EV) for public transport is being explored in Iloilo province in response to rising oil prices and environmental concerns. This study focuses on optimizing the location of the charging stations, considering the limitations of the EV driving range. The objective is to identify a minimal number of charging stations while ensuring that each node in the network is within a distance defined by traffic density-weighted conditions from a charging station. Data for the Iloilo Province road network were extracted using the Open Street Map package. The optimization problem was formulated and solved using the CPLEX method through the Gurobi Optimization Solver. Three formulations were introduced to address the EV Charging Stations Location (EVCSL) Problem, with the Set Covering Formulation being preferred due to its fewer variables. These formulations were applied to subnetworks extracted from Iloilo province. Secondary objectives, such as maximizing multiply-covered nodes and maximizing the minimum distance between charging stations, were also considered. The entire Iloilo network consists of 21,366 nodes, the node-to-node distance matrix cannot be obtained, and the EVCSL problem is not solved. Consequently, network reduction techniques were applied, which includes the removal of low-degree nodes and the consolidation of intersections. Results indicated that 39 charging stations are required for the reduced network using intersection consolidation, and 36 charging stations for the reduced network using the removal of low-degree nodes. The implemented codes extend the study’s applicability beyond Iloilo province to other regions. This is a joint work with Renier G. Mendoza (Computational Science Research Center, University of the Philippines Diliman, Philippines) and Maria Brenda Rayco (Institute of Mathematics, University of the Philippines Diliman, Philippines)."

TBA

In the Philippines, flooding poses great risks to life and property, with heavy rainfalls occurring more frequently in recent years. Effective mobilization is crucial to mitigate the severe consequences of flood events. However, studies on optimal routes that consider flood hazard levels in the country remain limited. In this study, we develop a novel flood route optimization model for the Philippines by integrating road network data from OpenStreetMap and flood hazard data from Project NOAH. The model is represented as a weighted graphG = (V, E, w), where the edge weights account for both travel distance and exposure to low, medium, and high flood hazard levels. We employed Non-dominated Sorting Genetic Algorithm II (NSGA-II) hybridized with Dijkstra’s algorithm to explore feasible routes that minimize the flood hazard exposure and the total distance. Applying our method to two flood-prone urban areas in Metro Manila—Project 4, Quezon City and Concepcion Dos, Marikina City—showed that the hybrid NSGA-II can generate a set of multiple solutions with diverse trade-offs between flood risk and distance, providing various options for decision-makers. This study highlights the importance of accounting for hazard exposure in flood route planning and using heuristic optimization methods in developing flood mobilization strategies.

We used persistent homology to detect topological signatures of admixture in 26 human populations consisting of 3,202 individuals from whole genome sequencing of the 1000 Genomes Project. We found that the topological structures, visualized as barcodes, can differentiate admixed populations from non-admixed populations. We also found that clustering using Wasserstein distance groups admixed populations together.

In this paper, we provide well-posedness results for nonlinear parabolic PDEs given by reaction diffusion equations describing the concentration of oxygen in encapsulated cells. The cells are described in terms of a core and a shell, which introduces a discontinuous diffusion coefficient as the material properties of the core and shell differ. In addition, the cells are subject to general nonlinear consumption of oxygen. As no monotonicity condition is imposed on the consumption monotone operator theory cannot be used. Moreover, the discontinuity in the diffusion coefficient bars us to apply classical results on strong solutions. However, by directly applying a Galerkin method we obtain uniqueness and existence of the strong form solution. These results will provide the basis to study the dynamics of cells in critical states.

TBA

The carabao (water buffalo) industry remains a vital component of the Philippine agricultural sector, providing draft power, meat, milk, and socio-economic security to smallholder farmers. In response to the stagnant carabao population, the Department of Agriculture (DA) launched a genetic upgrading program through artificial insemination (AI) to improve herd quality and productivity. However, calf production success remains limited. This study presents a predictive model for calf production outcomes in artificially inseminated carabaos, utilizing five years of national data collected from twelve regional centers of the Philippine Carabao Center (PCC). Correlation analysis identified the type of estrus (natural vs. estrus synchronization) and the distance between AI service centers and farm locations as significant predictors of successful calf production. Land-based distances were computed using OpenStreetMap tools (OSMnx), while inter-island distances were estimated using the OSMNx land distances and the Haversine formula. Machine learning models were trained using an 80:20 train-test data split, with XGBoost achieving the highest performance in terms of accuracy, area under the curve (AUC), and F1 score among all models tested. In addition, a decision-support dashboard was developed to provide real-time predictions of success rates compared to the national average, enabling more informed decision-making. The findings offer valuable insights into the geographic and biological factors influencing AI success in carabaos and support the development of more targeted, data-driven, and cost-effective strategies for improving reproductive outcomes in the Philippine livestock sector.

TBA