by Dr. Ohmini Krishnamurthy Rajendran*
ABSTRACT
Combating cancer is a significant challenge. Numerous cancer treatments are available. The medications have worked better. Making cancer medicines costs a lot of money and takes a long time. The recycling of medicines is proposed as a solution to these problems using computational methods like deep and automatic learning. For the purpose of repositioning cancer medications and predicting responses, traditional methods of automatic learning have been surpassed by deep learning. This study aims to develop a deep learning model that can anticipate responses to anticancer medications by making use of multi-omic and medication-related data in order to facilitate medication repositioning. Multi-omic data dimensionality is decreased by autoencoders. Polyvalent. Autoencoders related to MLPs To determine how effective our model is, we tested it on GDSC, CTRP, and CCLE. In many instances, our model always performs better than the current methods. Our model has an AUPRC of 0,99, which is significantly higher than that of other models. With an AUPRC of 0,72, the model developed on GDSC and tested on CCLE surpasses three previous studies. In conclusion, our deep learning model surpasses the current models. Utilizing this model, we intend to test drug responses and investigate reprogramming in an effort to discover novel cancer treatments. Our research demonstrates that sophisticated deep learning can improve anticancer treatments’ precision.
