Você está na página 1de 2

INTRODUCTION

CO-CLUSTERING
 Also called as bi-cluster is defined
as a submatrix spanned by a set
of genes and a set of samples.
 co-clustering simultaneously
clusters both samples and
features (genes)

1
INTRODUCTION
PROBLEM BACKGROUND
 Many researchers are continuously seeking for accurate
clustering algorithms for the creation of patient gene expression
profiles (Le Cao et al., 2011; Paz and Seeberger, 2012).

 It is challenging to analyze the genomic data in order to cluster


cancer patients of a certain histological or pathological cancer
type into different molecular subgroups (subtypes) (Ma et al.,
2014)

 Identifying cancer subtypes is a crucial task for selecting the right


treatment for patients, as different cancer subtypes may respond
well to different treatment therapies. (Xu et al., 2016)

 gene expression profiles generated very high dimension, i.e. the


number of genes is far larger than the number of samples lead to
dimensionality problem. ((Khashei et al., 2012; Tomasev et al.,
2014; Liu et. Al 2014)
 including too many variables may decrease the accuracy in clustering the samples, and
make the cluster rules difficult to set. The inclusion of irrelevant or noisy variables may
also degrade the overall performances of the estimated cluster rules.

 most previous studies did not incorporate biological information,


particularly molecular interactions networks, into the clustering
step (Liu et. Al 2014).

Você também pode gostar