Machine Learning-based epitope-mediated nanoMIPs as intelligent recognition materials for virus detection

• Applicant: Professorin Dr. Zeynep Altintas
• Subject Area: Analytical Chemistry ; Synthesis and Properties of Functional Materials
• Term: since 2025
• Project identifier: Deutsche Forschungsgemeinschaft (DFG) - Project number 557801359

This project highlights the significant potential of advancements in virus detection technology, driven by the integration of advanced sensing platforms with smart, receptor like bioinspired materials, which are essential for the precise and accurate identification of viral compounds. The proposed approach begins with establishing an extensive and meticulously curated library of epitopes, carefully extracted from the proteinous coats of multiple viruses. This epitope library forms the foundational basis for the design of molecularly imprinted polymers (MIPs)-based materials, which are tailored using advanced and extensive computational methods, ensuring that MIPs are specifically engineered to mimic the natural binding sites of viral molecules, thereby enhancing their specificity and effectiveness in virus detection. To further refine the design process of these materials, cutting-edge machine learning (ML) models are employed, as they play a pivotal role in streamlining the materials design process, allowing for the rapid and efficient optimization of MIPs. By analyzing vast, hybrid datasets and learning from molecular interactions, the developed ML model will enhance the precision with which MIPs are designed, ensuring that they achieve the highest possible accuracy in recognizing and binding to their target viral compounds. The integration of ML into the design process not only accelerates the development of smart receptors, but also significantly improves their binding performance, making them more reliable and effective in real-world applications. Following the design and optimization, the most promising MIP recipes will be synthesized, and afterwards subjected to rigorous testing through biosensing assays, specifically established to evaluate the (re)binding performance of MIPs generated through comprehensive molecular simulations and ML-driven optimization. The best-performing MIPs are then selected for integration into an ELISA platform, leveraging the enhanced selectivity and sensitivity of these advanced materials. The sensing platforms of interest are engineered for highly specific virus detection, with the ability to differentiate between closely related viral structures. Our proposed approach, based on the integration of optimized MIP components into biosensor and ELISA platforms represents a significant leap forward in virus detection. As a result, this project offers a robust solution to the existing challenges in the field, and promises a profound impact on public health by enabling faster, more accurate, and more reliable detection of viral infections, thereby contributing to better disease management and control.

• DFG Programme: Research Grants