Drug discovery is a complex and time-consuming process that involves fitting chemical compounds with proteins in the body to produce therapeutic effects. To expedite this process, researchers at SMU have developed SmartCADD, an open-source virtual tool that combines artificial intelligence, quantum mechanics, and Computer Assisted Drug Design techniques. In a recent study published in the Journal of Chemical Information and Modeling, SmartCADD demonstrated its ability to identify promising HIV drug candidates, significantly reducing drug discovery timelines. This tool emerged from an interdisciplinary collaboration between SMU’s department of chemistry and computer science department.
SmartCADD works by combining deep learning models, filtering processes, and explainable AI to screen databases of chemical compounds for potential drug leads. It has two main components – the Pipeline Interface that collects data and runs filters, and the Filter Interface that dictates how each filter should function. These built-in filters help predict how a drug will behave in the body, model drug structures using 2D and 3D parameters, and utilize an AI model that explains its decisions. The tool was demonstrated through three case studies focusing on drugs used to treat HIV, illustrating its versatility and adaptability to various drug discovery pipelines.
Interdisciplinary collaboration at SMU played a crucial role in the development of SmartCADD. Authors of the study include chemistry postdoctoral research fellow Ayesh Madushanka and computer science graduate student Eli Laird, among others. The strength of interdisciplinary collaboration was emphasized by the researchers, stating that fields like drug discovery require a combined effort for success. By bringing fresh perspectives and refining ideas, interdisciplinary collaboration can lead to significant advancements in research. Funding for the study was provided by the National Science Foundation, allowing researchers to continue developing and expanding the capabilities of SmartCADD.
The urgency to discover new classes of drugs like antibiotics, cancer treatments, and antivirals has driven the adoption of AI in scientific research. SmartCADD addresses concerns about AI’s opaqueness and data quality by sifting through billions of chemical compounds in a day, which significantly reduces the time needed to identify promising drug candidates. The tool is user-friendly and provides researchers with a highly integrated and flexible framework for building drug discovery pipelines. Researchers remain dedicated to pushing the project forward to enhance chemistry and machine learning capabilities and to make even more significant advancements.
While the study focused on HIV targets, SmartCADD’s versatility allows it to be applied to various drug discovery pipelines beyond HIV research. The Collaborative effort of researchers from different fields has proven the effectiveness of interdisciplinary research in making significant advancements that impact the real world. The interdisciplinary nature of the research at SMU reflects the university’s focus on fostering breakthrough innovations at the intersection of different fields. The researchers aim to position their work at the forefront of impactful research by continuing to collaborate across disciplines and spark ideas that lead to true innovations.
