Recently, his group proposed that solutions of pharmaceutical ingredients might behave as equilibrated molecules and associated colloids. They use the singular value decomposition (SVD) to analyze observed spectra-matrices, dividing them into basis vectors of spectra (a set of finite spectral elements shared by the individual observed spectra), singular values (corresponding to the intensity/standard deviation of each basis vector’s contribution), and singular vectors (a set of coefficient vectors for the linear combination of the basis vectors rearranged in the observed spectra). These singular vectors can reveal the configuration space, reflecting the constitution of the observed spectra (UV-vis, FTIR, fluoresence, CD, XRPD, ESR, and DSC) along experimentally conditional variables and visualizing their interconversion pathways depending on gradual changes in the observed or latent variables. In other words, the SVD treatment of the obtained instrumental data not only aids in trend analysis and noise reduction but also has the potential to uncover scientific laws in spontaneous phenomena.

有機化学では分子は亀の甲で表します。量子化学では原子核と電子雲の集合体です。けれども、幾何学的な図形なのだと思って構造式を見ていると、医薬品分子がまるで器械体操をしているかのようにダイナミックに動きまわる様子が見えてくるのです。その動きがあるからこそ、生物活性があり、そして医薬として治療効果があるのかもしれません。コンピュータやNMRやX線結晶解析でそんな研究に取り組んでいます。

1. アミロイド染色性色素の溶液における構造形成とこれに伴う疎水性の変動

2. 局所麻酔剤とNSAIDsの混合による共融混合物形成とその溶解性改善効果

3. 人工細胞膜リポソームを用いた薬物の生体膜透過性モデルの確立