Abstract: Index for water quality is solely based upon the conditions of physical and chemical which provide quantitative data on the presence and level of aquatic pollution. However, the parameters for the assessment do not represent the environmental stress and ecological health of the river which affect the microbial diversity. In this study, we introduced a new approach of assessing river water from metagenomic view by constructing phenetic tree which show relation between microbial community and the level of pollution. Four river waters, of which one river is categorised as unpolluted water, located around Johor Bahru (Malaysia) were examined and compared. Genomic DNA of these uncultured community of microorganisms was directly extracted and 16S rRNA gene was amplified by PCR using a set of primer pairs to generate cloned libraries. In total 24 isolates which are consisted of 18 isolates from polluted rivers and six from unpolluted one, were sequenced. Along these sequences, six of 16S rRNA sequences of colliform bacteria obtained from genebank were placed in Operational Taxonomic Units (OTUs). Phenetic analysis revealed that the river water used in this study were classified into two groups, representing the polluted and unpolluted. Since the tree can clearly distinct unpolluted river water from that of polluted one, we tried to develop a putative sequences motif for both condition. In practical purpose, the sequences motif can be used to screen river water quality in certain river water environment.