University of Houston

Genomic differences (mutations) in humans are profoundly influenced by their distinction as either germ line (inherited) or somatic (developed over one's life span). Such mutations can vary from a single nucleotide insertion, deletion, or substitution in a gene to a complete duplication or deletion of a large amount of genomic material ranging from thousands of nucleotides to an entire chromosome ultimately referred to as Copy Number Variations (CNV). While a large number of genomic variations have no significant influence on the overall quality of life, certain types of variations in a human genome called abnormalities are known to be associated with genetic disorders including cancer, autism, schizophrenia, just to name a few. Recent advancements in DNA sequencing technologies have made it possible to utilize High Throughput Sequencing (HTS) for the identification and detection of CNVs. The focus of this research is on the methods used to address the computational challenges associated with the application of high throughput sequencing for the detection of copy number variations in relatively large genomes (e.g. human genome) including the data representation and quality assessment. An evolutionary programming approach has been developed to use the set of novel algorithms and data structures introduced in this manuscript for the purpose of efficiently and accurately mapping genomic reads to one or more reference genomes. The developed computational tools have made it possible to identify the undesirable effects of repetitive regions in the human genome with the ability to identify CNVs and propose a novel approach to reduce them.