Genetic Signal Processing for Categorizing Genomic Data using Convolutional Neural Networks

Breakthroughs in high-throughput sequencing technology have resulted in exponential growth in acquired genetic data. Classic machine learning systems have difficulties in dealing with complex genetic data, leading to investigation of deep learning methods. A unique technique for categorizing genomic data using convolutional neural networks (CNNs) is described. The genetic signal processing approach can extract unique characteristics from genomic data. CNNs' ability to comprehend subtle patterns and correlations in cleaned-up data results in improved classification accuracy. A publicly accessible dataset is used to compare the proposed technique across various criteria. The study of genetic information using digital noise processing technology is known as genetic signal processing (GSP). GSP are used to turn the nucleotide sequence into a line graph, allowing for appropriate pattern classification. From the results, it clearly shows that proposed CNN achieves better results in terms of accuracy (0.96), precision (0.96), sensitivity (0.95), specificity (0.97) and f1-score (0.95).