Cloud Computing for Drug Discovery and Telemedicine in QSAR Analysis

This research is focused on developing drug discovery and telemedicine through advanced Quantitative Structure-Activity Relationship (QSAR) analysis. The performance of state-of-the-art machine learning (ML) algorithms like Random Forest (RF), DNN-QSAR, Support Vector Machine (SVM), was evaluated and the potential of Chemoinformatic based feature selection methods were explored for increasing the accuracy of developed QSAR models. However, not much innovation has been proposed in managing large chemical data and developing highly generalizable and more accurate QSAR models specific to telemedicine with low computational cost. Therefore, to fill this gap, we intended to use cloud computing (CC) as a downscale solution to host large chemical data and develop highly generalizable QSAR models that are more appropriate for telemedicine. This new way integrating cloud with QSAR for telemedicine is compatible with the world of current scientific research in pharmaceutical industries in which constant evolution and updating is a routine process. In addition to all these features integration of cloud also overcomes several inherited limitations of being non-performing with respect to speed up/QSAR analysis time or limited number or size chemicals database limitation that are used only in academic research studies.