Fig. 4From: PWSC: a novel clustering method based on polynomial weight-adjusted sparse clustering for sparse biomedical data and its application in cancer subtypingSubtyping of colorectal cancer cohort using tumor microbial data. a The consensus clustering CDF curve when k is from 2 to 15. b The consensus matrix when k is 4. c The clustering heatmap of microbial data for five groups. d The most valuable biomarkers with high occurrence in each group. e The Entropy of PWSC and conventional algorithm when k is from 2 to 15. f The CHI of PWSC and conventional algorithm when k is from 2 to 15Back to article page