the separation of the eight components was attained by using this LC fingerprint analysis method. For calculation approach to characteristics of LC Docetaxel Microtubule Formation inhibitor fingerprints of 11 source Dhge. Like a sort of TCM isatidis, there were two algorithms usually used: one was the correlation coefficient method, and the other was the cosine price method of vectorial angle. The treatments are as follows: where Xi is the peak area or peak height corresponding to the retention time in one sample, Yi is the peak area or peak height corresponding to the retention time in the reference fingerprint, X is the average peak area or peak height in this examined sample, Y is the average peak area or peak height in the reference fingerprint, d is the number of common peaks. The Similarity Evaluation System was useful for analyzing similarities of different chromatograms by calculating the correlation coefficients, in the same PTM time, other types of similarities of these chromatograms were also calculated on application of own modified Microsoft Excel system program based on the cosine value method of vectorial angle. The result of the characteristics of 11 Dtc. isatidis chromatograms is shown in Table 3. Good consistence was shown by the result obtained from the two algorithms together in development although there have been some differences in some places. After LC fingerprint fitting by variable wavelength mix technique and data analyses, the simulative mean chromatogram as a representative standard fingerprint of those R. isatidis samples from 11 sources was determined and created, and the reference fingerprinting profile is shown in Fig. 3B, showing large peak locations and good separation from surrounding mountains. The total peak areas of 24 common mountains were more than 80% of the total peak areas. 3. 4 HCA As discussed above, the information pan Aurora Kinase inhibitor listed in Table 3 unmasked variations in similarities between different sources. It would therefore be of interest to see if the sample set could be further divided into subgroups according to HCA. HCA is a statistical method to find reasonably homogeneous clusters of cases in accordance with measured characteristics, there are two main kinds of for HCA comprising agglomerative and divisive that find clusters of observations within a data set. The start with all the findings in a single bunch and then proceed to partition them into smaller clusters. The agglomerative begin with each observation being considered as split up clusters and then go to combine them until all observations belong to one cluster. On each stage, the set of groups with smallest cluster to cluster distance is fused into a single cluster. Used, the agglomerative were of greater use, so the agglomerative were chosen here whose result was represented graphically like a dendrogram.