Quantitative Structure Activity Relationship modeling and prediction is the fundamental methodology in an indirect drug design approach and is integral to a rational drug design process. QSAR modeling and analysis of results involves multiple steps requiring in depth knowledge of not only chemistry but also of statistics. The correct interpretation of results from a QSAR model is a key to utility of the entire QSAR exercise.
VLife QSAR Modeling services can assist at every step of a QSAR process beginning from advising on the choice of QSAR method to application of the results. Our scientific team has significant expertise in developing and applying QSAR techniques for new compound discovery as well as for optimizing discovered compounds. VLife’s patent pending GQSAR methodology enables our scientific team to accurately derive guidelines for modifying properties on a compound at specific sites. This provides an unprecedented advantage as it allows use of generated QSAR models to not only rationalize search results from chemical databases but to actually design or optimize new molecules.
QSAR is necessarily an iterative process and an in depth knowledge of the method and statistics is required to ensure that the iterations minimized. VLife’s QSAR Modeling services typically involve activities such as descriptor choice and calculation, statistical evaluation of the calculated descriptors, training and test set assignment, regression and results analysis. Our scientists evaluate multiple options for classes of descriptors, assignment of training and test set, choice of linear or non linear regression and choice of regression technique to determine the option that is most suitable to a particular project. They exercise an intelligent choice of parameters during each of these stages to deliver the most accurate and robust QSAR models in a time efficient manner. The QSAR Modeling services thus facilitate a complete ligand based design approach.