Genome analysis is an important aspect of understanding the underlying meaning of the gene sequencing data in bioinformatics. Different machine learning techniques have been implemented to predict if the drug sensitivity is high or low depending on IC50 values. After testing several algorithms such as random forest, Naive Bayes, CatBoost, gradient boosting, and neural networks, the study provides an accuracy of 99.81% using the CatBoost algorithm and 99.28% accuracy using neural network, which is the highest. The performance evaluation for each one was done using a confusion matrix, F1 score, precision, and recall. To improve model performance, extensive data preprocessing and feature visualization approaches are utilized from the Genomics of Drug Sensitivity in Cancer (GDSC) dataset containing the genomic information of cancer cells and the drug response. The findings show that deep learning approaches and ensemble methods, especially CatBoost, are useful for classifying drug sensitivity and providing information on the genetic markers linked to cancer therapy response.