Sample-distance partial least squares: PLS optimized for many variables, with application to CoMFA
| Autor: | Bruce L. Bush, Robert B. Nachbar |
|---|---|
| Rok vydání: | 1993 |
| Předmět: |
Models
Molecular Computer science Fortran Histamine Antagonists In Vitro Techniques Machine learning computer.software_genre Field (computer science) Matrix (mathematics) Resampling Drug Discovery Partial least squares regression Humans Computer Simulation Least-Squares Analysis Physical and Theoretical Chemistry Topology (chemistry) computer.programming_language Transcortin Molecular Structure Covariance matrix business.industry Computer Science Applications Steroids Pairwise comparison Artificial intelligence business computer Algorithm Software |
| Zdroj: | Journal of Computer-Aided Molecular Design. 7:587-619 |
| ISSN: | 1573-4951 0920-654X |
| Popis: | Three-dimensional molecular modeling can provide an unlimited number m of structural properties. Comparative Molecular Field Analysis (CoMFA), for example, may calculate thousands of field values for each model structure. When m is large, partial least squares (PLS) is the statistical method of choice for fitting and predicting biological responses. Yet PLS is usually implemented in a property-based fashion which is optimal only for small m. We describe here a sample-based formulation of PLS which can be used to fit any single response (bioactivity). SAMPLS reduces all explanatory data to the pairwise 'distances' among n samples (molecules), or equivalently to an n-by-n covariance matrix C. This matrix, unmodified, can be used to fit all PLS components. Furthermore, SAMPLS will validate the model by modern resampling techniques, at a cost independent of m. We have implemented SAMPLS as a Fortran program and have reproduced conventional and cross-validated PLS analyses of data from two published studies. Full (leave-each-out) cross-validation of a typical CoMFA takes 0.2 CPU s. SAMPLS is thus ideally suited to structure-activity analysis based on CoMFA fields or bonded topology. The sample-distance formulation also relates PLS to methods like cluster analysis and nonlinear mapping, and shows how drastically PLS simplifies the information in CoMFA fields. |
| Databáze: | OpenAIRE |
| Externí odkaz: |
|
Full text from SpringerLink