Chapter 8 - Computational genomics for understanding of DNA-DNA and protein-protein similarity

Abstract

Whole-genome mapping has been made possible by bacterial enzymatic DNA fragmentation, strand separation, and sequencing in several organisms. There have been advances in next-generation genomic sequencing (NGS) with the application of poly-oligonucleotide functionalized nanotechnology for high-throughput re-sequencing. Predictive proteomic analysis and transcriptional genomic application algorithms have been developed in parallel to advances in lexicographic suffix and sub-prefix indexing with iterarive partitioning for improved accuracy in query with scaled compression in linearization of computation, and compression algorithms. Computational nomics begins with development of algorithms for global and local sequence alignment, amino acid sequence homology detection in a substitution matrix design, oligonucleotide mutation affinity studies for position weight matrix-based design, and then extends into sequence dis-similarity detection with improved accuracy. NGS nanotechnologies result in high-throughput RNA sequencing for gene expression transcriptomics, DNA sequencing (or exonic RNA) for high sensitivity and specificity probability polymorphism detection and mutational mapping alignment genomics that have been in parallel to advances in computational algorithms for low-error sequencing, alignment and variant calling re-alignment, and open-access data science repositories. Since data sampling meets the normal tendency expectation, there exist various probability distribution models that have been computationally-applied, which include those such as Poisson, half-Gamma, Gumbel, or Dirichlet, in addition to models with a priori association–related variables such as the Bayesian, internodal chain, and expectation-maximization with iterative maximum likelihood. In this chapter, the computational algorithms and applied methodologies for genomic mapping and assembly, re-sequencing reads alignment, tagged gene retrieval, transcription factor binding to DNA response elements, binding site clustering, RNA splice form likelihood are discussed with perspective on the future applicability of computational advances in combinatorial analyses for proteomics, genomics, and transcriptome modeling in the macromolecular probability for inhomology relationship.