Nansu Zong

This paper evaluates the use of GPT for drug sensitivity prediction using pharmacogenomics data, showing improved performance through prompt engineering and fine-tuning.

The paper introduces a drug-target prediction method that improves accuracy and scalability by integrating heterogeneous networks, Node2Vec for feature learning, and a hybrid model with negative sampling, achieving high AUC ROC scores.

The paper introduces DATS, a metadata model for dataset discoverability, and describes its implementation for the DataMed data discovery index.

DataMed enhances data discoverability by creating a unified index of biomedical datasets across repositories, supporting FAIR principles for better knowledge discovery.