Comparative Analysis of Machine Learning Algorithms for DNA Sequencing

A genome consists of the genetic information of the organism. Sequencing millions of DNA molecules in a human individual provides the information about the genetic makeup to the doctors. This technique is essential for identifying the variants and mutations that contribute to a particular genetic disorder. Researchers and doctors use this piece of information to better understand the patient's current state of health and to aid in the development of treatment and operations in the future. DNA sequencing is done to track epidemics, learn about the past portfolio of the individual and develop better treatments for life threatening diseases like cancer and arthritis. Understanding a patient's unique genetic profile can allow drug makers to target specific subgroups of individuals grouped together by similar genetic makeup. This may lead to more precise and customized information, such as the type of drug being used and the dosage level. Hence DNA sequencing plays an integral role in the biomedical research process. In this work we analyze different machine learning techniques like SVM, CNN, LSTM, Random forest classifier, Adaboost, Naive bayes, Multilayer perceptron, XGB Classifier to sequence the DNA on a human dataset. This study analysis can aid us to understand which algorithm works better with satisfying accuracy.