Transformer-based large language models (LLMs) have revolutionized the ways to understand and explore massive datasets and enable us efficiently generate and augment relevant data. Now, these transformer models have extended applicability in understanding languages of the molecules of life — DNA, RNA, proteins, and chemical compounds. Understanding and exploring such vast and intricate chemical and biological data spaces efficiently are crucial for expediting drug discovery. BioNeMo — built using NVIDIA’s NeMo framework — enables training of transformer-based LLMs and their use for inference. We'll introduce BioNeMo and its applications, then take a deeper dive into the inference using pre-trained models available with BioNeMo — ESM1nv, ProtT5nv, and MegaMolBART, with a walk-through of chemical space exploration and biological inference. We'll also provide an example of using the learned embeddings from the pre-trained model for property predictions.
Prerequisite(s):
Please disregard any reference to "Event Code" for access to training materials. "Event Codes" are only valid during the original live session. Explore more training options offered by the NVIDIA Deep Learning Institute (DLI). Choose from an extensive catalog of self-paced, online courses or instructor-led virtual workshops to help you develop key skills in AI, HPC, graphics & simulation, and more.