An Inverse QSAR Method Based on Linear Regression and Integer Programming
Background: Drug design is one of the important applications of biological science. Extensive studies have been done on computer-aided drug design based on inverse quantitative structure activity relationship (inverse QSAR), which is to infer chemical compounds from given chemical activities and con...
| Main Authors: | , , , , , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
IMR Press
2022-06-01
|
| Series: | Frontiers in Bioscience-Landmark |
| Subjects: | |
| Online Access: | https://www.imrpress.com/journal/FBL/27/6/10.31083/j.fbl2706188 |
| _version_ | 1818255717755256832 |
|---|---|
| author | Jianshen Zhu Naveed Ahmed Azam Kazuya Haraguchi Liang Zhao Hiroshi Nagamochi Tatsuya Akutsu |
| author_facet | Jianshen Zhu Naveed Ahmed Azam Kazuya Haraguchi Liang Zhao Hiroshi Nagamochi Tatsuya Akutsu |
| author_sort | Jianshen Zhu |
| collection | DOAJ |
| description | Background: Drug design is one of the important applications of biological science. Extensive studies have been done on computer-aided drug design based on inverse quantitative structure activity relationship (inverse QSAR), which is to infer chemical compounds from given chemical activities and constraints. However, exact or optimal solutions are not guaranteed in most of the existing methods. Method: Recently a novel framework based on artificial neural networks (ANNs) and mixed integer linear programming (MILP) has been proposed for designing chemical structures. This framework consists of two phases: an ANN is used to construct a prediction function, and then an MILP formulated on the trained ANN and a graph search algorithm are used to infer desired chemical structures. In this paper, we use linear regression instead of ANNs to construct a prediction function. For this, we derive a novel MILP formulation that simulates the computation process of a prediction function by linear regression. Results: For the first phase, we performed computational experiments using 18 chemical properties, and the proposed method achieved good prediction accuracy for a relatively large number of properties, in comparison with ANNs in our previous work. For the second phase, we performed computational experiments on five chemical properties, and the method could infer chemical structures with around up to 50 non-hydrogen atoms. Conclusions: Combination of linear regression and integer programming is a potentially useful approach to computational molecular design. |
| first_indexed | 2024-12-12T17:16:18Z |
| format | Article |
| id | doaj.art-c4d5b2d8c8264ab3baac7b57322ebbd0 |
| institution | Directory Open Access Journal |
| issn | 2768-6701 |
| language | English |
| last_indexed | 2024-12-12T17:16:18Z |
| publishDate | 2022-06-01 |
| publisher | IMR Press |
| record_format | Article |
| series | Frontiers in Bioscience-Landmark |
| spelling | doaj.art-c4d5b2d8c8264ab3baac7b57322ebbd02022-12-22T00:17:46ZengIMR PressFrontiers in Bioscience-Landmark2768-67012022-06-0127618810.31083/j.fbl2706188S2768-6701(22)00551-2An Inverse QSAR Method Based on Linear Regression and Integer ProgrammingJianshen Zhu0Naveed Ahmed Azam1Kazuya Haraguchi2Liang Zhao3Hiroshi Nagamochi4Tatsuya Akutsu5Department of Applied Mathematics and Physics, Kyoto University, 606-8501 Kyoto, JapanDepartment of Applied Mathematics and Physics, Kyoto University, 606-8501 Kyoto, JapanDepartment of Applied Mathematics and Physics, Kyoto University, 606-8501 Kyoto, JapanGraduate School of Advanced Integrated Studies in Human Survavibility (Shishu-Kan), Kyoto University, 606-8306 Kyoto, JapanDepartment of Applied Mathematics and Physics, Kyoto University, 606-8501 Kyoto, JapanBioinformatics Center, Institute for Chemical Research, Kyoto University, 611-0011 Uji, JapanBackground: Drug design is one of the important applications of biological science. Extensive studies have been done on computer-aided drug design based on inverse quantitative structure activity relationship (inverse QSAR), which is to infer chemical compounds from given chemical activities and constraints. However, exact or optimal solutions are not guaranteed in most of the existing methods. Method: Recently a novel framework based on artificial neural networks (ANNs) and mixed integer linear programming (MILP) has been proposed for designing chemical structures. This framework consists of two phases: an ANN is used to construct a prediction function, and then an MILP formulated on the trained ANN and a graph search algorithm are used to infer desired chemical structures. In this paper, we use linear regression instead of ANNs to construct a prediction function. For this, we derive a novel MILP formulation that simulates the computation process of a prediction function by linear regression. Results: For the first phase, we performed computational experiments using 18 chemical properties, and the proposed method achieved good prediction accuracy for a relatively large number of properties, in comparison with ANNs in our previous work. For the second phase, we performed computational experiments on five chemical properties, and the method could infer chemical structures with around up to 50 non-hydrogen atoms. Conclusions: Combination of linear regression and integer programming is a potentially useful approach to computational molecular design.https://www.imrpress.com/journal/FBL/27/6/10.31083/j.fbl2706188machine learninglinear regressioninteger programmingchemoinformaticsmaterials informaticsqsar/qsprmolecular design |
| spellingShingle | Jianshen Zhu Naveed Ahmed Azam Kazuya Haraguchi Liang Zhao Hiroshi Nagamochi Tatsuya Akutsu An Inverse QSAR Method Based on Linear Regression and Integer Programming Frontiers in Bioscience-Landmark machine learning linear regression integer programming chemoinformatics materials informatics qsar/qspr molecular design |
| title | An Inverse QSAR Method Based on Linear Regression and Integer Programming |
| title_full | An Inverse QSAR Method Based on Linear Regression and Integer Programming |
| title_fullStr | An Inverse QSAR Method Based on Linear Regression and Integer Programming |
| title_full_unstemmed | An Inverse QSAR Method Based on Linear Regression and Integer Programming |
| title_short | An Inverse QSAR Method Based on Linear Regression and Integer Programming |
| title_sort | inverse qsar method based on linear regression and integer programming |
| topic | machine learning linear regression integer programming chemoinformatics materials informatics qsar/qspr molecular design |
| url | https://www.imrpress.com/journal/FBL/27/6/10.31083/j.fbl2706188 |
| work_keys_str_mv | AT jianshenzhu aninverseqsarmethodbasedonlinearregressionandintegerprogramming AT naveedahmedazam aninverseqsarmethodbasedonlinearregressionandintegerprogramming AT kazuyaharaguchi aninverseqsarmethodbasedonlinearregressionandintegerprogramming AT liangzhao aninverseqsarmethodbasedonlinearregressionandintegerprogramming AT hiroshinagamochi aninverseqsarmethodbasedonlinearregressionandintegerprogramming AT tatsuyaakutsu aninverseqsarmethodbasedonlinearregressionandintegerprogramming AT jianshenzhu inverseqsarmethodbasedonlinearregressionandintegerprogramming AT naveedahmedazam inverseqsarmethodbasedonlinearregressionandintegerprogramming AT kazuyaharaguchi inverseqsarmethodbasedonlinearregressionandintegerprogramming AT liangzhao inverseqsarmethodbasedonlinearregressionandintegerprogramming AT hiroshinagamochi inverseqsarmethodbasedonlinearregressionandintegerprogramming AT tatsuyaakutsu inverseqsarmethodbasedonlinearregressionandintegerprogramming |