Browsing by Author "Chandrasekaran, Sunita"
Now showing 1 - 2 of 2
Results Per Page
Sort Options
Item Machine learning classifier approaches for predicting response to RTK-type-III inhibitors demonstrate high accuracy using transcriptomic signatures and ex vivo data(Bioinformatics Advances, 2023-03-24) Ferrato, Mauricio H.; Marsh, Adam G.; Franke, Karl R.; Huang, Benjamin J.; Kolb, E. Anders; DeRyckere, Deborah; Grahm, Douglas K.; Chandrasekaran, Sunita; Crowgey, Erin L.Motivation: The application of machine learning (ML) techniques in the medical field has demonstrated both successes and challenges in the precision medicine era. The ability to accurately classify a subject as a potential responder versus a nonresponder to a given therapy is still an active area of research pushing the field to create new approaches for applying machine-learning techniques. In this study, we leveraged publicly available data through the BeatAML initiative. Specifically, we used gene count data, generated via RNA-seq, from 451 individuals matched with ex vivo data generated from treatment with RTK-type-III inhibitors. Three feature selection techniques were tested, principal component analysis, Shapley Additive Explanation (SHAP) technique and differential gene expression analysis, with three different classifiers, XGBoost, LightGBM and random forest (RF). Sensitivity versus specificity was analyzed using the area under the curve (AUC)-receiver operating curves (ROCs) for every model developed. Results: Our work demonstrated that feature selection technique, rather than the classifier, had the greatest impact on model performance. The SHAP technique outperformed the other feature selection techniques and was able to with high accuracy predict outcome response, with the highest performing model: Foretinib with 89% AUC using the SHAP technique and RF classifier. Our ML pipelines demonstrate that at the time of diagnosis, a transcriptomics signature exists that can potentially predict response to treatment, demonstrating the potential of using ML applications in precision medicine efforts. Availability and implementation: https://github.com/UD-CRPL/RCDML Supplementary information: Supplementary data are available at Bioinformatics Advances online at: https://doi.org/10.1093/bioadv/vbad034Item Proceedings of the 2023 Delaware Data Science Symposium(Data Science Institute of the University of Delaware, 2023-09-22) Bagozzi, Benjamin E.; Abou Ali, Hanan; Blaustein, Michael; Blinova, Daria; Buler, Jeffrey; Carney, Lynette; Chandrasekaran, Sunita; Davey, Adam; Fleischhacker, Adam; Ostovari, Mina; Peart, Daniel; Smith, Sam; Tawiah, Nii Adjetey; Wu, Cathy H.The 2023 Delaware Data Science Symposium was held on September 22nd with a primary focus on the role of data science in financial technology (FinTech) and health equity. The Symposium was organized by the University of Delaware’s (UD’s) Data Science Institute (DSI) with support from Tech Impact, Dupont, Kendal Corporation, Intellitec Solutions, UD’s Library, Museums, & Press, the UD Career Center, the UD Graduate College, the UD Master of Science in Data Science Program, UD’s Artificial Intelligence Center of Excellence (AICOE), and the DSI. It represented the fourth Delaware Data Science Symposium hosted at the University of Delaware, and the third such Symposium since the DSI’s inception. Altogether, the Symposium saw over 280 registered attendees from the University of Delaware and partner institutions across the Mid-Atlantic and beyond. The 2023 Delaware Data Science Symposium included multiple keynote speakers, a series of initiative & lightning talks, a poster session, a panel on data science-driven equity from healthcare, FinTech, community, and educational perspectives, and a session on UD’s summer 2023 Data Science (DS) + Artificial Intelligence (AI) Hackathon. Alongside these sessions, the Symposium also facilitated two associated satellite events. The first was a September 21st Data Science and Analytics Open House for UD graduate programs focused on data science and analytics. The second was a September 25th workshop on the use of MATLAB for low-code AI.