Traditional genomic analysis and identification of patterns and regularities within genomic sequences is a cornerstone of bioinformatics that frequently relies on sequence alignment, a process that is often computationally intensive and requires precise positional anchoring. This thesis explores alternative, alignment-free methodologies to identify structural patterns and functional regions within DNA sequences. The core of the research involves the application of specialized algorithms to compute similarity and complexity across sequences. Specifically, we utilize kwip (k-mer weighted inner product) for alignment-free distance estimation and Kullback-Leibler (KL) divergence to quantify the entropic differences between genomic regions. This approach is designed to distinguish functional genes from non-functional sequences by identifying unique informational signatures.