Skip to main content
. 2015 Aug 12;5:12894. doi: 10.1038/srep12894

Figure 3.

Figure 3

Robustness of Progeny Clustering against varied sampling sizes (5, 10 and 20 progenies generated per cluster) in the (A) two-dimensional and (B) ten-dimensional toy datasets. In both cases, the three curves derived using different sampling sizes were consistent in trends, which indicates the algorithm’s robustness. (C) Sensitivity of Progeny Clustering to noise present in the three-cluster, two-dimensional dataset. Noise was generated with σ ranging from 0.1 to 1.5, and the frequency of correct estimation was calculated based on ten repeated experiments.