Fast dimension reduction and integrative clustering of multi-omics data using low-rank approximation: application to cancer molecular classification

Table 2 The results of single-cancer analysis

Cancer	Dimension^a	#Cluster^b	Silhouette values
BRCA	2	2	0.55
COAD	4	4	0.40
GBM	8	2	0.35
HNSC	7	3	0.26
KIRC	6	2	0.36
LGG	2	3	0.44
LUAD	5	2	0.34
LUSC	5	4	0.32
PRAD	2	4	0.41
STAD	4	3	0.37
THCA	2	2	0.61

^aThe dimension of the reduced space is determined according to the curve of the explained variations of each cancer type
^bThe number of clusters is determined according to the curve of the within cluster variances

ISSN: 1471-2164