ResearchTrend.AI
  • Papers
  • Communities
  • Events
  • Blog
  • Pricing
Papers
Communities
Social Events
Terms and Conditions
Pricing
Parameter LabParameter LabTwitterGitHubLinkedInBlueskyYoutube

© 2025 ResearchTrend.AI, All rights reserved.

  1. Home
  2. Papers
  3. 2107.02381
18
1

An Inverse QSAR Method Based on Linear Regression and Integer Programming

6 July 2021
Jianshen Zhu
Naveed Ahmed Azam
Kazuya Haraguchi
Liang Zhao
H. Nagamochi
Tatsuya Akutsu
ArXivPDFHTML
Abstract

Recently a novel framework has been proposed for designing the molecular structure of chemical compounds using both artificial neural networks (ANNs) and mixed integer linear programming (MILP). In the framework, we first define a feature vector f(C)f(C)f(C) of a chemical graph CCC and construct an ANN that maps x=f(C)x=f(C)x=f(C) to a predicted value η(x)\eta(x)η(x) of a chemical property π\piπ to CCC. After this, we formulate an MILP that simulates the computation process of f(C)f(C)f(C) from CCC and that of η(x)\eta(x)η(x) from xxx. Given a target value y∗y^*y∗ of the chemical property π\piπ, we infer a chemical graph C†C^\daggerC† such that η(f(C†))=y∗\eta(f(C^\dagger))=y^*η(f(C†))=y∗ by solving the MILP. In this paper, we use linear regression to construct a prediction function η\etaη instead of ANNs. For this, we derive an MILP formulation that simulates the computation process of a prediction function by linear regression. The results of computational experiments suggest our method can infer chemical graphs with around up to 50 non-hydrogen atoms.

View on arXiv
Comments on this paper