A Novel QPAR Technique for Extracting Valuable Information from Herbal Medicines

This novel quantitative-pattern-activity-relationship (QPAR) technique brings the breakthrough to answer two significant existing problems in the studying of complex active mixtures: (1) Can we predict accurately the total functional activities of unknown sample based on only the chromatographic profile? and (2) Can we identify those features in the chromatographic fingerprint responsible for such activities? In studying two herbal medicines, the results obtained clearly demonstrated the ability of the technique in solving these two problems. Its prediction capability is better than 92% in the total immunological enhancement bioactivity of one of them. Another breakthrough is successful prediction of the antioxidant active fractions of the other one without the need of separating individual fractions through experimentation.