Figure 9

An illustration of phenotype tree construction process. (a) Images 1–25, 26–45, 46–61, and 62–77 correspond to non-neoplastic S1 cells cultured for 12 days, 10 days, 5 days, and 3 days respectively. There are 7 possible ways of grouping the phenotypes. Each row corresponds to one possible way. Different colors represent different phenotype groups. The first 3 rows correspond to grouping the 4 predefined phenotypes into 2 groups. The next 3 rows correspond to grouping the phenotypes into 3 groups, and the last row correspond to 4 groups. (b) Taking the 4 phenotype group case (last row in (a)) as an example, we used traditional clustering methods to divide the cluster histogram of the image (one cluster histogram per image) into the same number of clusters (i.e., 4 in this example). Each row corresponds to the clustering result of one method. (c) The F-measures computed by pairing the phenotype group in the last row of (a) with each clustering result in (b). The maximum F-score, which in this case is achieved by the Gaussian Mixture Model approach (GM), is selected as the confidence of the corresponding cell phenotype grouping. (d) Confidence values as functions of different cases of phenotype groupings. We tested the confidence values under different number of clusters predefined for clustering LBF distributions using the five traditional methods (i.e., the second step of our algorithm, see Figure 6) as shown by dots of different colors. The numbers of clusters we tested were 4 to 26 with step size of 2. The consistent distribution of the dots indicates that our phenotype tree construction method is insensitive to the number of clusters we selected for clustering LBF distributions.