Regularized regression analysis of digitized molecular structures in organic reactions for quantification of steric effects

Abstract

In organic chemistry, Comparative Molecular Field Analysis (CoMFA) can be defined as a regression analysis between reaction outcomes and molecular fields, wherein we can extract and visualize important structural information from the coefficients of the constructed regression models. In CoMFA, partial least-squares (PLS) regression, which determines all coefficients in the model, is used for fitting the regression models. However, in organic reactions, steric effects are observed only near the reactive site, indicating that a large number of regression coefficients in the CoMFA of organic reactions should be assigned as 0. The regularized regression method, LASSO/Elastic Net, allows us to fit the regression model while assigning 0 values to unimportant coefficients. Although LASSO/Elastic Net should be suitable for CoMFA, there is no example of its use for organic reaction analysis. Herein, we examine the performance of LASSO/Elastic Net for the quantification of steric effects in CoMFA. We employ digitized molecular structures (the indicator field) as molecular fields that represent steric effects. LASSO/Elastic Net regressions provide highly interpretable models that include less noise than those from PLS regression. © 2017 Wiley Periodicals, Inc.

Data-driven catalyst optimization for stereodivergent asymmetric synthesis of α-allyl carboxylic acids by iridium/boron hybrid catalysis

“…1d and see Supplementary fig. S1) were employed, which are often used for MFA in asymmetric catalysis 4,40,[49][50][51] . It has been recognized that weak attractive interactions, such as dispersion forces, are sometimes important for asymmetric catalysis [52][53][54][55] .…”

“…As target variables, logarithms of constitutional and diastereomeric ratios (b/l and dr) (G ‡ = -RTlog(b/l or dr)) were employed, which correspond to energy differences between the transition states that lead to each isomer (Curtin-Hammett principle 56 ). By correlating the target variables and the molecular fields through LASSO 57 or Elastic Net 58 using the R package, glmnet 59 according to the reported procedure 51 , four regression models were generated for predicting b/l and dr in the reactions using the S or R ligand.…”

Descriptors for dielectric constants of perovskite-type oxides by materials informatics with first-principles density functional theory

“…the so-called quantitative structure-property relationships (QSPRs), between the structures and properties of functional materials [10]. In this case, regression analysis methods, such as partial least-squares (PLS) regression, principal component analysis (PCA), support vector regression (SVR), and least absolute shrinkage and selection operator (LASSO) regression, are often used for building regression models for molecules and crystal structures [11][12][13][14][15][16][17][18][19]. Clarifying the descriptors indicating the QSPRs will accelerate the design of functional materials.…”

Molecular Field Analysis Using Computational-Screening Data in Asymmetric N-Heterocyclic Carbene-Copper Catalysis toward Data-Driven In Silico Catalyst Optimization

“…The indicator fields were calculated from the TS structures of a minor pathway (the R pathway), as we aimed to obtain guidelines for the design of chiral NHC ligands that would destabilize the minor pathway. By correlating the calculated DDG ‡ values and the indicator fields using LASSO 13 or Elastic Net 14 regression according to the reported procedure 15 employing the R package glmnet 16 , we generated a regression model (for more details regarding the MFA, see the SI section 'Details of the molecular field analysis'). The important structural information obtained from the regression coefficients is visualized in the TS structures of L1S1 and L3S1 (Figure 3a and 3b).…”