月別: 2026年1月

[Published online Journal of Computer Chemistry, Japan Vol.24, 80-82, by J-STAGE]
<Title:> SVDを用いた実時間TDDFTのスペクトル解析
<Author(s):> 谷　直樹, 狩野　覚, 善甫　康成
<Corresponding author E-Mill:> naoki.tani.7x(at)stu.hosei.ac.jp
<Abstract:> Optical spectrum prediction based on first-principles calculations is important for the development of optical materials. In particular, Time Dependent Density Functional Theory (TDDFT) in real-time is one of the most widely used calculation methods. In real-time TDDFT, the dynamic dipole moment is used to obtain the polarizability by Fourier transform (FT). The optical spectrum can be obtained from this polarizability. However, if the time length is not sufficient, the spectrum resolution depends on the length and becomes ambiguous. To solve this problem, we introduced Dynamic Mode Decomposition (DMD). This spectral analysis technique uses both Singular Value Decomposition and Proper Orthogonal Decomposition to obtain frequencies and intensities of spectrum. The signal frequency and the intensity are directly obtained. This method was applied to TDDFT time series data for ethylene and small molecules of benzene, naphthalene, anthracene, and tetracene. Compared to conventional FT, clear spectra were obtained from a short time series data.
<Keywords:> Keywords TDDFT, Dynamic Mode Decomposition (DMD), Fourier Transform (FT), Singular Value Decomposition (SVD), Real-time TDDFT
<URL:> https://www.jstage.jst.go.jp/article/jccj/24/3/24_2025-0010/_article/-char/ja/

[Published online Journal of Computer Chemistry, Japan Vol.24, 86-89, by J-STAGE]
<Title:> 親水性カーボンナノオニオンの溶媒抽出法に対する分子動力学計算による検討
<Author(s):> 守屋　祐輝, YEH　Jungwen, 内田　正哉
<Corresponding author E-Mill:> f5005ddf(at)sit.ac.jp
<Abstract:> Carbon nano-onions (CNOs) are concentric fullerene-like shells recognized for stability and versatility. We evaluated the liquid liquid extraction of hydroxylated CNOs in a toluene water system using molecular dynamics simulations. The analysis of trajectories, radial distribution functions (RDFs), and spatial distribution functions (SDFs) revealed that increasing the surface OH density decreases the coordination number of toluene, redistributes toluene molecules to non-functionalized patches, and alters their spatial distribution. Highly hydroxylated CNOs tended to reside in the water phase or at the interface. Collectively, these results identify surface OH density as the key determinant of solvent selectivity and phase distribution, enabling practical control of CNO partitioning during liquid liquid extraction.
<Keywords:> キーワード molecular dynamics, carbon nano-onions, hydroxylation, toluene water extraction, RDF, SDF.
<URL:> https://www.jstage.jst.go.jp/article/jccj/24/3/24_2025-0019/_article/-char/ja/

[Published online Journal of Computer Chemistry, Japan Vol.24, 83-85, by J-STAGE]
<Title:> Physics-Based Docking Complements AlphaFold-Multimer Based Selection in de novo Binder Design: A HER2 Affibody Case Study
<Author(s):> Masatomo GO, Motoki INOUE, Takefumi YAMASHITA
<Corresponding author E-Mill:> yamashita.takefumi(at)hoshi.ac.jp
<Abstract:> Accurate antigen binder interfaces are critical for de novo campaigns that use AlphaFold-Multimer (AFM) in selection. In the HER2 ZHER2 affibody case, AFM produced correct monomer folds but placed the binder at a non-native site. Inter-chain Predicted Aligned Error indicated low interface confidence for the AFM complexes. By contrast, a physics-based workflow combining ZDOCK, RMSD-based clustering, and HADDOCK refinement recovered near-native interfaces. These observations indicate that physics-based docking may serve as a complementary approach to AFM-based selection.
<Keywords:> De novo binder design, AlphaFold-Multimer, Docking, Affibody, HER2
<URL:> https://www.jstage.jst.go.jp/article/jccj/24/3/24_2025-0013/_article/-char/ja/