Biosensors Conference

Abstract

Non-coding RNAs, mainly microRNAs and long non-coding RNAs, have emerged as key regulators in cancer. These orchestrate epigenetic modifications and chromatin remodeling to modulate transcriptional programs implicated in tumorigenesis. Dysregulation of certain miRNAs or lncRNAs can promote oncogenesis and metastasis; for example, the loss of tumor-suppressor miR-34a or overexpression of oncogenic lncRNAs such as HOTAIR and MALAT1 facilitates malignant progression.

miRNAs and lncRNAs also contribute to the cancer immunomodulatory landscape. miRNAs can promote anti-tumor immunity or facilitate immune evasion by reprogramming immune cells and cytokine networks within the tumor microenvironment; some miRNAs boost cytotoxic T cell responses while others expand immunosuppressive regulatory T cells and myeloid-derived suppressor cells. They also regulate immune checkpoint molecules, including PD-L1 and CTLA-4, thereby influencing tumor immunogenicity and response to immunotherapy. Likewise, lncRNAs modulate antigen presentation and immune cell differentiation, thus tipping the balance between pro- and anti-tumor immune responses. Certain lncRNAs induce T-regs and M2 macrophages to create an immunosuppressive microenvironment that enables tumor immune escape.

Given the diverse regulatory roles of miRNAs and lncRNAs, both classes of non-coding RNAs are being intensively investigated as potential biomarkers and therapeutic targets in cancer. Their expression profiles are closely associated with patient prognosis and can serve as predictors of response to immunotherapy. Therapeutically, oligonucleotide-based interventions are being developed to modulate aberrant ncRNA activity—miRNA mimics and antagomirs aim to restore tumor-suppressive miRNAs or inhibit oncogenic miRNAs, whereas siRNAs are utilized to silence oncogenic lncRNAs. Early clinical studies have demonstrated encouraging outcomes, though challenges persist, particularly in achieving tumor-specific delivery while minimizing off-target effects. Advances in chemical modification and nanoparticle-mediated delivery systems are progressively addressing these limitations. A comprehensive overview of these strategies will be presented during the conference. Continued exploration of ncRNA-mediated epigenetic and immunoregulatory networks holds significant promise for the development of novel biomarkers and precision-based cancer therapeutics, despite existing translational challenges.

Keywords: Non-coding RNAs; microRNAs; long non-coding RNAs; cancer immunotherapy; epigenetic regulation.

Abstract

Recent advances in biosensing demand analytical techniques that can interpret rapidly changing biological signals with high precision, noise immunity, and transparency. This work introduces Dynamic Lissajous Topography (DLT), a new geometric approach for representing and analyzing biosignals in real time. Instead of relying on traditional frequency-domain or statistical transformations, DLT models the interaction between biosensor outputs—such as optical intensity versus phase, enzyme potential versus current, or neural voltage versus impedance—as evolving geometric trajectories in a dynamic space. In this framework, each biosensing pair forms a time-varying Lissajous pattern whose shape continuously deforms in response to biochemical or physiological changes. These deformations capture intrinsic information about reaction kinetics, molecular binding events, or physiological states. The curvature and topological evolution of the trajectory serve as indicators of system stability, synchronization, and anomaly occurrence. Laboratory simulations and prototype biosensor data reveal that DLT enables instantaneous recognition of subtle biochemical changes with exceptional resilience to drift and noise. It allows real-time visualization of biological processes, adaptive anomaly localization, and interpretable differentiation between normal and pathological patterns. The geometric descriptors can be implemented efficiently on embedded hardware or edge devices, supporting portable and wearable biosensing applications. Dynamic Lissajous Topography establishes a fundamentally new perspective on biosignal interpretation—one that views biological interactions as evolving geometric forms rather than static signals. This geometry-driven methodology offers a promising pathway toward intelligent, low-power, and explainable biosensing systems capable of adaptive, continuous health monitoring.

Abstract

Optical metamaterials are artificial structures designed to precisely manipulate electromagnetic waves. Among them, hyperbolic metamaterials (HMMs) stand out due to their high anisotropy and hyperbolic dispersion relation, enabling applications in negative refraction, high-resolution imaging, and biosensing. Specifically, the integration of HMMs into Surface Plasmon Resonance (SPR) biosensors has enhanced key performance parameters such as sensitivity, quality factor, figure of merit, and detection accuracy, due to the improved confinement of the electromagnetic field at the biosensor/analyte interface. However, predicting performance parameters is challenging because of the variety of metrics involved in SPR biosensors. In this context, artificial intelligence (AI) has emerged as a powerful tool for forecasting performance parameters. In this work, we generate a dataset of SPR biosensors with a Kretschmann-type configuration (prism/metal layer/HMM/analyte) using the Transfer Matrix Method (TMM). This dataset served as the input for training various machine learning models, including linear regression, polynomial regression, Support Vector Machine (SVM), and Random Forest, to predict performance parameters such as sensitivity and figure of merit, which were defined as target outputs. The predicted results showed strong agreement with computational values obtained via the TMM. Finally, the models were validated using techniques such as cross-validation and error minimization metrics, with Random Forest delivering the most accurate predictions.

Abstract

Modern life increasingly relies on advanced technologies that provide real-time information to support decision-making in both daily and industrial contexts. These technologies are largely based on sensors of various types, including chemical, mechanical, pressure, acoustic, and optical sensors. Among them, surface plasmon resonance (SPR) sensors have proven highly effective for detecting a wide range of analytes, such as ions, proteins, viruses, and small molecules. This versatility enables applications in environmental monitoring, water and food quality control, industrial safety, and clinical diagnostics. The development of two-dimensional (2D) materials, such as graphene, hexagonal boron nitride, black phosphorene, and MXenes, has significantly enhanced the performance of conventional SPR sensors due to their remarkable optical properties. Furthermore, diabetes mellitus remains a growing global health concern, currently affecting over 537 million people. Projections by the International Diabetes Federation (IDF) estimate that this number could reach 700 million by 2045. Diabetes involves complex metabolic disorders and is characterized by elevated blood glucose levels, which can lead to severe complications such as kidney disease, limb amputation, and premature death. This project focuses on improving the performance of SPR biosensors for glucose detection. We analyze prism-coupled SPR biosensors in the Kretschmann configuration, composed of heterostructures made from metals, dielectrics, and 2D materials. The system is evaluated using attenuated total reflection (ATR), calculated via the transfer matrix method. Our theoretical results demonstrate a 30% increase in sensitivity and approximately 40% improvement in both the quality factor and detection accuracy, highlighting the potential of 2D- material-based SPR sensors for biomedical applications.