Bishnu Sarker, Ph.D., assistant professor of computer science and data science, recently led a tutorial at ISMB/ECCB 2023, one of the premier conferences in the field of computational biology.
Dr. Sarker, along with colleague Sayane Shome, Ph.D., a postdoctoral researcher at Stanford University, presented the virtual session on Protein Sequence Analysis using Transformer-based Large Language Model. They guided 60 participants through experiential training on how to build basic machine learning pipelines using deep learning and pre-trained transformer protein language models for biological sequence analysis.
The tutorial covered an important topic for research in both computational science and biology.
“Artificial intelligence and machine learning have tremendously facilitated scientific discoveries in biomedicine,” says Dr. Sarker.
“Recent advancements in the development of large language models, a type of deep learning model that can read, summarize, translate, and generate text as we humans do, have inspired many researchers to find applications in biological sequence analysis, partly because of the similarities in the data,” he says.
“For example,” he adds, “ProtTrans was trained on large dataset of UniProtKB protein sequences. They showed that transformer-based, self-supervised protein language models effectively capture the long-distance spatial relationship among residues, which is critical for understanding the functional and structural aspects of proteins.”
Though Dr. Sarker has previously attended several sessions and presented papers at the ISMB/ECCB conference, this was his first time leading a tutorial. It was an invaluable experience for both him and Meharry’s students.
“Preparing a tutorial for an audience of diverse academic and professional backgrounds surely improves the classroom experience for Meharry SACS’s students,” says Dr. Sarker. “They will greatly benefit from the materials, like the slides and Python practice notes, that are already provided in the tutorial webpage as well as from the GitHub repository. “
Of course, he also appreciates the honor of being selected for one of the most prestigious conferences in his field.
“Presenting a tutorial in such a top conference is always an honor and privilege,” he says. “The fact that our proposal successfully passed such a competitive review is an acknowledgement of our merit and expertise.”
“I would like to thank the Dean Dr. Fortune Mhlanga and the Chair Dr. Tim Wallace for supporting these professional development activities and personally encouraging me to present this tutorial,” adds Dr. Sarker.