|
Affiliation |
Faculty of Science Department of Mathematics and Infomation Science |
|
Title |
Professor |
|
External Link |
|
TERAMAE Hiroyuki
|
|
|
Research Areas 【 display / non-display 】
-
Nanotechnology/Materials / Fundamental physical chemistry
From School 【 display / non-display 】
-
Kyoto University Faculty of Engineering Graduated
- 1979.03
Country：Japan
From Graduate School 【 display / non-display 】
-
Kyoto University Graduate School, Division of Engineering Doctor's Course Completed
- 1984.03
Country：Japan
-
Kyoto University Graduate School, Division of Engineering Master's Course Completed
- 1981.03
Country：Japan
Employment Record in Research 【 display / non-display 】
-
Josai University Abolition organization Depertment of Chemistry Professor
2004.04
External Career 【 display / non-display 】
-
Meiji University Lecturer
2009.04 - 2023.03
Country：Japan
-
Rikkyo University Lecturer
2002.04 - 2006.03
-
ATR適応コミュニケーション研究所 主任研究員
2001.10 - 2002.03
-
ATR環境適応通信研究所 主任研究員
1999.01 - 2001.09
-
Rikkyo University Lecturer
1997.04 - 1998.03
Professional Memberships 【 display / non-display 】
-
日本コンピュータ化学会
1998.04
-
アメリカ化学会
1996.04
-
応用物理学会
1992.07 - 1996.03
-
日本化学会
1990.04
Research Career 【 display / non-display 】
-
分子軌道法と機械学習による分子物性の予測
The Other Research Programs
Project Year： 2017.04 -
-
分子軌道法を用いたナノマテリアル内での化学反応に関する研究
Funded Research
Project Year： 2008.12 - 2014.03
-
分子内プロトン移動反応に関する理論的研究
Cooperative Research
Project Year： 2008.04 - 2013.03
-
高次元アルゴリズムによる分子構造最適化の研究
The Other Research Programs
Project Year： 2004.04 -
Papers 【 display / non-display 】
-
Prediction of pharmacological activity by machine learning using molecular orbital energy as an explanatory variable Invited Reviewed
23 ( 3 ) 80 - 83 2025.01
Authorship：Lead author, Corresponding author Language：Japanese Publishing type：Research paper (scientific journal)
We constructed a mathematical model to predict the 2,2-diphenyl-1-picrylhydrazyl (DPPH) free radical scavenging capacity (IC50) for recently synthesized ferulic acid derivatives by machine learning with molecular orbital energy as an explanatory variable and IC50 as an objective variable. We compared 96 regression models including xgbLinear and neuralnet included in R/caret package. We were able to construct IC50 prediction models for these new ferulic acids by using xgbLinear, M5, ppr, and neuralnet as regression methods.
-
Prediction of log P Parameter Using Molecular Orbital Energies and Machine Learning Invited Reviewed
Hiroyuki Teramae
22 ( 2 ) 34 - 36 2024.02
Authorship：Lead author, Corresponding author Language：Japanese Publishing type：Research paper (scientific journal)
Octanol/water partition coefficient, log P, is an important parameter in classical QSAR. The new method using machine learning which we propose uses only the molecular orbital energy as an explanatory variable and does not include log P. Therefore, since the log P value can be predicted using the molecular orbital energy, we speculated that log P may not be necessary as a result if sufficient number of molecular orbital energies would be given as parameters.
-
Prediction of Entropy by Machine Learning with Molecular Orbital Energies Invited Reviewed
Takafumi Yuuki, Wakana Nakahara, Hiroyuki Teramae
22 ( 2 ) 31 - 33 2024.02
Authorship：Corresponding author Language：Japanese Publishing type：Research paper (scientific journal)
The values of the entropy of 148 small organic molecules have been estimated by machine learning with only molecular orbital energies as the explanatory variables. Out of 148 molecules, we used 104 molecules for the training set and 44 molecules for the test set. We used 139 regression methods of R/caret packege for machine learning. We evaluated values by RMSE (Root Mean Squared Error) and R² (coefficient of determination). From those evaluation, xgbLinear (eXtreme Gradient Boosting) and RRFglobal (Regularized Random Forest) are considered better than other regression methods. It has been proved that the entropy can be predicted by the molecular orbital energies only.
-
Machine Learning Study of Antioxidant Effects with Molecular Orbital Energies as Explanatory Variables Invited Reviewed
Hiroyuki Teramae, Meiyan Xuan, Jun Takayama, Mari Okazaki, Takeshi Sakamoto
21 ( 4 ) 103 - 105 2023.04
Authorship：Lead author, Corresponding author Language：Japanese Publishing type：Research paper (scientific journal)
The values of the internuclear distances and the dipole moments of 14 small molecules have been estimated by machine learning with only molecular orbital energies as the explanatory variables. We use four regression methods, partial least square (PLS), random forest (RF), Radial Basis Function Kernel Regularized Least Squares (krlsRadial), and Baysian Regularized Neural Networks (BRNN) and we report only BRNN results for the internuclear distances, and PLS results for the dipole moments. The coefficients of determination for the internulear distances and the dipole moments are 0.9318 and 0.7265, respectively. It has been proved that the internuclear distances and the dipole moments can be predicted by the molecular orbital energies only.
-
Prediction of molecular properties with machine learning and molecular orbital energies Invited Reviewed
Hiroyuki Teramae, Meiyan Xuan, Jun Takayama, Mari Okazaki and Takeshi Sakamoto
AIP Conference Proceedints 2611 02007 2022.11
Authorship：Lead author, Corresponding author Language：English Publishing type：Research paper (international conference proceedings)
The prediction by the machine learning using molecular orbital energies as an explanatory variable is attempted to predict the strength of anxiolytics, anti-anxiety, and muscle relaxant of benzodiazepine anxiolytics. We also attempt to predict half-life of concentration in the body T1/2, and time to reach maximum body concentration Tmax of benzodiazepine anxiolytics with the same procedure. The molecular orbital calculations are performed at 6-31G(d, p) level and random forest is used as regression method. The number of molecular orbitals is varied from 2 to 20 and it is found that 4 or 6 is almost sufficient for the prediction of these 5 objective variables. Finally, the predictions of five properties in the present study are fairly well agreed with the experiments by machine learning employing the molecular orbital energies as the only explanatory variables.
Books and Other Publications 【 display / non-display 】
-
ケモインフォマティクスにおける データ収集の最適化と解析手法
寺前裕之（ Role： Contributor , 第4章第1節ケモインフォマティクスにおける機械学習モデルの種類と具体的活用法）
技術情報協会 2023.04 （ ISBN:978-4-86104-944-6 ）
Total pages：657 Responsible for pages：209-217 Language：Japanese Book type：Scholarly book
-
Materials Informatics Questions and Answers
Masahiro Kaneko, Kimito Funatsu, Hiroyuki Teramae etc.（ Role： Joint author , Chapter 8, Section 3, Question 4）
JOHOKIKO CO. LTD. 2020.12 （ ISBN:978-4-86502-204-9 ）
Total pages：597 Responsible for pages：516-518 Language：Japanese Book type：Scholarly book
A number of questions that arise during the introduction and operation of materials informatics are specifically resolved in a Q & A format.
-
導電性材料をめぐる最近の動向
寺前裕之（ Role： Sole author）
材料技術研究協会 1992.04
Language：Japanese Book type：Scholarly book
-
ポリアセチレンの電子構造
山邊時雄，寺前裕之（ Role： Sole author , 主要部分の執筆）
化学同人 1985.04
Language：Japanese Book type：Scholarly book
Misc 【 display / non-display 】
-
Relation Between Machine Learning and Chemistry Invited
2022.03
Language：Japanese
-
PubChem のデータを用いた分子座標の作成‐データベースを用いてGaussian16 の入力ファイルを作成する方法 ‐
寺前裕之
城西情報科学研究 29 15 - 26 2022.03
Authorship：Lead author, Last author Language：Japanese Publishing type：Rapid communication, short report, research note, etc. (bulletin of university, research institution)
-
計算化学汎用プログラム 分子設計統合ソフト HyperChem
寺前裕之
PETROTECH 30 ( 5 ) 346 - 350 2007.01
Language：Japanese Publishing type：Article, review, commentary, editorial, etc. (international conference proceedings) Publisher：（石油学会）
Presentations 【 display / non-display 】
-
Prediction of pharmacologicalactivity by machine learning usingmolecular orbital energy as anexplanatory variable International conference
Hiroyuki Teramae
ICCMSE2024 2024.09
Event date： 2024.09
Language：Japanese Presentation type：Oral presentation (invited, special)
Venue：Greece
-
分子軌道エネルギーと機械学習による新規フェルラ酸誘導体のラジカル消去能の予測
寺前 裕之, 三浦 優太, 色摩 光一,玄 美燕, 高山 淳, 岡﨑 真理, 坂本武史
第18回分子科学討論会（京都）
Event date： 2024.09
Language：Japanese Presentation type：Oral presentation (general)
Venue：京都
-
説明変数に分子軌道エネルギーのみを用いた機械学習によるエントロピーの予測
結城 敬史, 寺前 裕之
日本コンピューター化学会2024年春季年会
Event date： 2024.06
Language：Japanese Presentation type：Poster presentation
Venue：東京
-
機械学習による分子軌道エネルギーのみを説明変数としたエントロピーの予測
結城敬史、寺前裕之
日本コンピュータ化学会2023年秋季年会 2023.11 日本コンピュータ化学会
Event date： 2023.11
Language：Japanese Presentation type：Poster presentation
Venue：東京 Country：Japan
-
分子軌道エネルギーを用いた機械学習によるlogPの予測
寺前裕之
分子科学討論会2023 2023.09
Event date： 2023.09
Language：Japanese Presentation type：Oral presentation (general)
Venue：大阪 Country：Japan
Other research activities 【 display / non-display 】
-
Editorial board of Journal of Chemistry
2015.01 - 2018.08