Abstract

*Ronald Bartzatt

ABSTRACT

Algorithmic multiple regression is utilized for analysis of molecular properties and for the predictions of potential anti-EBOLA compounds. In this study, is the analysis of 20 tested anti-EBOLA agents, specifically their molecular properties, with multiple regression analysis for predicting the critical property of molecular weight of potential anti-EBOLA compounds. The molecular properties used are: Log P, molecular volume, polar surface area, and number of atoms. Using these molecular properties, multiple regression analysis successfully generated an equation having R2 of 0.9318, and therefore accounting for 93.18 % of the variance in the dependent variable of molecular weight. Descriptive statistics of molecular properties for these 20 anti- EBOLA compounds are presented. Application of prediction methods will assist in the development of novel pharmaceuticals, as contemporary techniques are well characterized and have been applied across a broad spectrum of disciplines. Pattern recognition techniques such as cluster analysis, 95% ellipses, and non-metric multidimensional scaling are applied to determine complex relationships found in this multivariate data, such as similarities (or dissimilarities). Descriptive statistics of molecular properties for these 20 anti- EBOLA compounds are presented.