الفهرس 
ContentsOnly 14 pages are availabe for public view
 |  | from | 100 |  |  |
| | | from | 100 | | |
Abstract
The thesis presents Third Generation Sequencing (TGS) and the most prominent companies that produce third generation sequencing devices, namely Pacific Biosciences (PacBio) and Oxford Nanopore Technology (ONT), with an explanation of how the third-generation sequencing devices work. Third generation sequencers (TGS) give long reads but with relatively high error rates. The thesis then reviews three quality-related metrics, basecalling accuracy, Phred quality, and GC content. The alternatives to the assembly process were also addressed, namely, overlap-layout-consensus (OLC), and de Bruijn graph (DBG). Also, deep neural networks were used for basecalling accuracy. The measured loss does not exceed 5.42. The thesis is divided into 5 chapters as follows : 1- Introduction: This chapter presents an introduction to the topic of the thesis in addition to the presentation of the problem statement, the objective of the thesis, and finally a brief presentation of the components of this thesis. 2- Previous Work: In this chapter, the relationship between DNA and proteins is presented. The applications of DNA sequencing in medicine, pharmaceutical production and forensic medicine are presented. Next generation sequencing (NGS) is also presented. Then, the chapter deals with Third Generation Sequencing (TGS). The main manufacturers of third-generation sequencing devices are Pacific Biosciences (PacBio) and Oxford Nanopore Technology (ONT) with an explanation of how the third-generation sequencing devices work. Third generation sequencers (TGS) give long reads but with relatively high error rates. The chapter then reviews three quality-related metrics, basecalling accuracy, Phred quality, and GC content. Finally, the chapter presents two methods overlap-layout-consensus (OLC) and de Bruijn graph (DBG).3- The proposed framework: Steps are proposed to explore the quality of third- generation sequencers, and deep neural networks have been used for basecalling accuracy. Finally, the chapter presents algorithms for the assembly alternatives: OLC and DBG. 4- Results and discussion: This chapter contains the results of exploring the quality of sequences. And for the accuracy of base calling, a deep neural network is adopted. The measured loss does not exceed 5.42. The chapter also reviews the results of the assembly alternatives OLC and DBG. 5- Conclusion and future actions: This chapter is about the conclusions obtained from the application and study of the proposed methodology and what can be added in the future to obtain better performance.