scispace - formally typeset
Journal ArticleDOI

On Some Aspects of Variable Selection for Partial Least Squares Regression Models

Partha Pratim Roy, +1 more
- 01 Mar 2008 - 
- Vol. 27, Iss: 3, pp 302-313
Reads0
Chats0
TLDR
In this article, the optimum variable selection strategy for Partial Least Squares (PLS) regression using a model dataset of cytoprotection data is explored, where the compounds of the dataset were classified using K-means clustering technique applied on standardized descriptor matrix and ten combinations of training and test sets were generated based on the obtained clusters.
Abstract
This paper tries to explore the optimum variable selection strategy for Partial Least Squares (PLS) regression using a model dataset of cytoprotection data. The compounds of the dataset were classified using K-means clustering technique applied on standardized descriptor matrix and ten combinations of training and test sets were generated based on the obtained clusters. For a particular training set, PLS models were developed with a number of components optimized by leave-one-out Q2 and then the developed models were validated (externally) using the test set compounds. For each set, PLS model was initially constructed using all descriptors (variables). The variables having least standardized values of regression coefficients were deleted and the next model was developed with a reduced set of variables. These steps were performed several times until further reduction in number of variables did not improve Q2 value. In each case, statistical parameters like predictive R2 (R2pred), squared correlation coefficient between observed and predicted values with (r2) and without () intercept and Root Mean Square Error of Prediction (RMSEP) were calculated from the test set compounds. In case of all ten sets, Q2 values steadily increase on deletion of variables while R2pred values do not show any specific trend. In no case, the highest Q2 and highest R2pred appear in the same trial, i.e., with the same combinations of variables. This suggests that from the viewpoint of external predictability, choice of variables for PLS based on Q2 value may not be optimum. Moreover, a clear separation of r2 and r02 curves in some sets suggests that such models may not be truly predictive in spite of acceptable R2pred values. Another observation is that coefficient of determination R2 for the training set is more immune to changes on deletion of variables than the validation parameters like Q2 and R2pred. Finally, a new parameter rm2 has been suggested to indicate external predictability of QSAR models.

read more

Citations
More filters
Journal ArticleDOI

A Comprehensive QSAR Study on Antileishmanial and Antitrypanosomal Cinnamate Ester Analogues.

TL;DR: In this article, a set of cinnamate esters and their antileishmanial and antitrypanosomal activity was investigated using a genetic algorithm as main variable selection tool and multiple linear regression as statistical analysis.
Journal ArticleDOI

A QSAR Study of Matrix Metalloproteinases Type 2 (MMP-2) Inhibitors with Cinnamoyl Pyrrolidine Derivatives.

TL;DR: A multivariate PLS-QSAR study with a data set of 31 cinnamoyl pyrrolidine derivatives described as type 2 matrix metalloproteinases (MMP-2) inhibitors is presented in this paper.
Journal ArticleDOI

MIA-QSAR modelling of activities of a series of AZT analogues: Bi- and multilinear PLS regression

TL;DR: The activities of a series of azidothymidine derivatives, compounds with anti-HIV potency, were computationally modelled using multivariate image analysis applied to quantitative structure–activity relationships (MIA-QSAR), and two regression methods were tested to find the best correlation between actual and predicted activities.
Journal ArticleDOI

Quantitative Structure–Property Relationship Study to Predict Speed of Sound in Diverse Organic Solvents from Solvent Structural Information

TL;DR: In this article, the effect of solvent structural parameters on the speed of sound in chemical solvents was investigated through a quantitative structure-property relationship (QSPR), which was employed to select the most relevant subset of descriptors and then to develop the model.
Journal ArticleDOI

Identifying the structural features and diversifying the chemical domain of peripherally acting CB1 receptor antagonists using molecular modeling techniques

TL;DR: In the present work, pharmacophore mapping studies on diverse classes of compounds such as imidazoles, purines, pyrazines, piperazines,pyrazoles and pyrazolines have been performed to get structural insights into the CB1 receptor antagonists having peripheral activity.
References
More filters
Book

Cluster Analysis

TL;DR: This fourth edition of the highly successful Cluster Analysis represents a thorough revision of the third edition and covers new and developing areas such as classification likelihood and neural networks for clustering.
Journal ArticleDOI

Beware of q2

TL;DR: It is argued that the high value of LOO q2 appears to be the necessary but not the sufficient condition for the model to have a high predictive power, which is the general property of QSAR models developed using LOO cross-validation.
Journal ArticleDOI

PLS regression methods

TL;DR: In this paper, the mathematical and statistical structure of PLS regression is developed and the PLS decomposition of the data matrices involved in model building is analyzed. But the PLP regression algorithm can be interpreted in a model building setting.
Related Papers (5)