Abstract
In recent years, the Next Generation Sequencing (NGS) Technologies have evolved rapidly and has led to various developments in genome assembly field. Denovo assembly is a core problem in bioinformatics and remains a complex task with the high throughput data produced by NGS sequencers. In this study, we present a new denovo assembly approach based on clustering. Our denovo assembly approach uses long reads generated by NGS sequencers and consists of two steps: during the first step, we create clusters based on reads similarity. Then during the second step, we apply an alignment to these reads grouped in every cluster to create contigs. To test the performance of our approach we use long reads generated by eukaryotic genomes and we compare the performance with assemblers widely used in denovo genomic assembly filed.