Optimization of bioinformatics software

• Main Author: Himaani Mahajan.
• Other Authors: Stephen John Turner
• Format: Final Year Project
• Language: English
• Published: 2012
• Subjects: DRNTU::Engineering::Computer science and engineering::Computer systems organization::Performance of systems
• Online Access: http://hdl.handle.net/10356/48792
• Institution: Nanyang Technological University

Increasing research is being done in various fields of bioinformatics like vaccine development, molecular engineering, manipulation of digital genetic coding etc. Artificial Gene or Genome Synthesis facilitates research in these areas, as custom designed DNA can be synthesised without the need for existing precursor template DNA. The process of gene synthesis uses text-based DNA sequences, specifying the exact order of nucleotide bases in a DNA, to synthesise a gene in vitro using standard assembly processes. Hence, gene synthesis is more economical than the classical cloning techniques. Many commercial software are available which provide services for gene synthesis, but they are limited by the length of DNA sequence they can synthesise. Increasingly long DNA sequences of the order of millions of base pairs are available, and the synthesis of such DNA sequences using the current systems is either not feasible or takes huge amounts of time making it impractical. Software for gene synthesis was developed by the author during Industrial Attachment at Institute of High Performance Computing (IHPC), which facilitates the synthesis of multi-million base pair long DNA sequences. However, the software could carry out the synthesis process for the longest DNA sequence tested (Escherichia Coli, ~5.5 million base pairs) in a very long time period of more than 6 hours. Hence, the focus of this project is to optimise the performance of the software using high performance computing techniques like parallelisation using OpenMP. Parallel computing enhances computational performance by executing multiple calculations (computations) simultaneously on different processing elements.