Clustering and Exploring of Gene Functional Modules from Cassava Root Gene Expression Data

Cassava is an important economic crop, both in Thailand and internationally. Advances in sequencing technology have allowed cassava genome to be deciphered. However, identifying the functions of all genes in the cassava genome using plant molecular biology laboratory requires is a tedious and resource-extensive. The present research therefore aimed to predict gene functions based on their expression profiles using the K-means clustering method and to propose their functions to unknown genes via the use of Gene Set Enrichment Analysis (GSEA). Three tissues of cassava roots, including storage root, fibrous root and root apical meristem, were used in the study. The gene expression data were divided into 2 subsets, which are SET1: fibrous root and root apical meristem and SET2: storage root, fibrous root and root apical meristem. Cassava genes could be divided into 21 groups and 20 groups, respectively; however, only 14 groups can be assigned the significant functions in both subsets. 8,561 and 8,727 unknown genes can be assigned the functions in SET1 and SET2, respectively. Totally, putative related functions can be assigned to 8,736 cassava genes or 26.45 percent of all the genes in the cassava genome. The results allow 75.38 percent of the genes in the genome to be assigned with their related functions.