OIST-JAIST Joint Symposium: "Collaborative Innovation for Sustainable Future through Advanced Materials Science"

The advancement of cancer research and treatment modalities has resulted in a reduction in mortality rates associated with numerous cancer types. However, the prognosis for fibrotic cancers, such as pancreatic cancer, remains poor, and no effective treatments have been developed. In collaboration with Kensuke Osada at the National Institute for Quantum Science and Technology, we have devised a novel strategy to enhance the efficacy of chemotherapy. This strategy involves the delivery of the anti-apoptotic gene-silencing siRNA to cancer cells using the uPIC nanoparticle, which is capable of penetrating the stroma. Pilot experiments have yielded promising results in a model with a highly developed stroma.

Dual-catalytic systems offer a transformative approach to facilitating previously inaccessible chemical reactions. By combining two distinct catalytic processes, these systems enable synergistic interactions that promote new reactivity pathways, leading to increased efficiency and selectivity in organic synthesis. In our group, we have been exploring the applicability of these reactions in semiconducting polymer syntheses and have been able to achieve controlled direct arylation polymerizations, room temperature reactions, and cross dehydrogenative reactions. These reactions could not be achieved using a single catalyst and highlight the utility of such systems. In this presentation, the achievements made will be discussed while also addressing the challenges that remain.

Materials informatics (MI) applies data science techniques to materials science, and its potential has been explored across numerous fields. However, successful application of data-driven approaches requires datasets that are sufficient in size, distribution, and consistency. A key challenge in MI is the scarcity of datasets suitable for machine learning. To address this, our research group employs high-throughput experimentation, from synthesis to evaluation, to generate high-quality and large-scale datasets. By integrating machine learning techniques, including advanced automatic feature engineering, we uncover robust hypotheses that significantly enhance the materials discovery process.