FY2024 The STAR Special Award for Science Tokyo Advanced Researchers Decided

Recently, the term “drug pricing” has been frequently mentioned in the news, highlighting the rising costs of healthcare. However, the development of a new drug typically takes more than 10 years and involves research and development expenses amounting to 20 billion dollars. Moreover, discovering completely new treatments requires exploring a wide range of possibilities and combining different approaches. In other words, drug discovery has become more complex, challenging, and expensive.
At the same time, advancements in computational technologies, such as AI, are progressing rapidly. A major challenge today is finding ways to apply these innovations to speed up and improve the drug discovery process. Current research in this field is focused on developing computational methods for designing new molecules. In recent years, the range of molecules—or modalities—in drug discovery, including small molecules, peptides, nucleotides, antibodies, and others. has grown significantly. This diversification calls for the development of tailored techniques for each modality.
For example, some approaches include free energy perturbation methods for small molecules (FastLomap, PairMap), explainable AI methods for predictions in molecular design (MMGX), and design techniques for antibody drugs that utilize tools like AlphaFold and protein language models. By using these methods, researchers are working to identify the most effective strategies for different modalities and to develop practical, ready-to-use technologies. Moreover, while each modality has its own unique features, a comprehensive, cross-disciplinary approach is believed to be key for the future advancement of drug discovery.

In the present day, humanity has reached an unprecedented level of control over electrons, spins, and photons in solids, leading to the creation of numerous innovative electronic devices. On the other hand, as the energy crisis becomes more severe, reducing and effectively utilizing the enormous waste heat generated by electronic devices and factories has become an important challenge. However, heat conduction in solids remains poorly understood, and the development of materials and technologies that enable advanced control and utilization of thermal energy is still a difficult task. To tackle this issue, we have developed material designs that enable the manipulation of electrons and phonons (atomic vibrations) through state-of-the-art computational science and experiments. We demonstrated eco-friendly high-performance thermoelectric materials that enable large-scale power generation from waste heat, metal-insulator junction materials that generate power using cryogenic heat, and heat management materials that can switch between thermal insulation and heat dissipation. Going forward, we will develop next-generation energy materials and devices for the advanced utilization of thermal energy.