Fig. 6
From: Detecting genomic deletions from high-throughput sequence data with unsupervised learning
Clustering results with unsupervised learning. A.1, B.1, C.1, D.1, E.1 The two axes are from the top two principle components of PCA. The dots represent all deletion candidates in chromosome 6, 10, 1, 4 and 13 of NA12777, NA12776, NA12878, NA12775 and NA12763 respectively. The cyan dots stand for the deletion candidates recorded in the 1000 Genomes Project Phase3 callset, which are viewed as true deletions. The black dots refer to the candidates that are not in Phase3 callset, which are viewed as false positives. A.2, B.2, C.2, D.2, E.2 Classification results of hierarchical clustering on chromosome 6, 10, 1, 4 and 13 of NA12777, NA12776, NA12878, NA12775 and NA12763 respectively. In each scatter plot, four clusters of deletions are classified, which are marked in different colors