Workshop 5 – Unfolding the Future: Protein Modeling, Targeted Drug Design, and Mutation Insights

Description

Abstract:

The understanding of protein structures has invaluable applications in the medical field, ranging from the identification of disease targets and design of novel lead molecules to characterizing underlying mechanisms of disease. Despite the major progress in genome sequencing, the
structural characterization of these proteins, which would enable their full exploration and exploitation, remains lagging. Fortunately, advancements in AI now permit the efficient and accurate prediction of protein structures through deep neural networks. This workshop, “Unfolding
the Future,” initially explores AI-powered techniques for highly accurate 3D protein predictions in an easy-to-follow protocol using python in Google Colab. After modeling the protein structure, participants will be guided through the process of docking, a fundamental technique used in drug design. Briefly, participants will initially learn to identify a ligand binding site within the protein and how to search for potential high affinity binders, which can be further optimized to modulate  protein function. The obtained protein/ligand complex will then be used for the final stage of the workshop, where participants will be guided into understanding the effects of missense mutations on protein structure and ligand affinity, which are central to its function. As missense mutations often lead to different genetic diseases, linking their clinical consequences to their effects on the protein can enable deeper insights into disease mechanisms than can be predicted using sequence alone. By the end of this workshop, participants will have a better appreciation of protein folding, function, and its application to the clinic.

Intended audience
This is a basic-level workshop designed for students and professionals with a biological background, with limited experience in structural biology, protein visualization and computational biology. Experience in scripting or coding is helpful, but not necessary for participation in this
workshop.

Synopsis
Advances in protein sequencing over the past few decades have helped identify proteins directly related to disease, however, the experimental elucidation of these proteins, which is crucial to their understanding, remains unoptimized. In this workshop, participants will get a step-by-step guide on how to model protein structure using AI, how to explore protein interactions with their respective ligand using docking analysis, and how to assess any changes inflicted to the protein structure and function upon addition of disease-causing mutations.

Materials
All the materials, including course slides, code snippets and worksheets will be provided during the workshop.

Biography

Yoochan Myung
Yoochan Myung is a research fellow at the University of Queensland, with honorary fellowships at Systems and Computational Biology at Bio21 Institute and Computational Biology and Clinical Informatics at the Baker Institute. His major research interests include antibody and vaccine design using machine learning and diverse computational biology approaches. (Twitter:@Yoochan_Myung)

Thanh-Binh Nguyen
Thanh-Binh Nguyen is a research fellow at the School of Chemistry and Molecular Biosciences (SCMB) at the University of Queensland (UQ). Her research interests lie in computational biology, especially protein structure modelling, virtual screening and molecular dynamics simulations. She also uses machine learning methods to study the effects of mutations on disease phenotypes. (Twitter: @BinhchemNguyen)

Stephanie Portelli
Stephanie Portelli is a trained Pharmacist and Computational Biologist with a strong interest in drug design, drug resistance and genetic disease. She obtained her PhD in Computational Biology and Clinical Informatics in 2021 from the University of Melbourne, and currently works as a research fellow within the School of Chemistry and Molecular Biosciences (SCMB) at the University of Queensland. Her research focuses on developing software to help doctors make more informed clinical decisions by studying the impact of pathogenic missense mutations on protein structure and function and incorporating that within a machine learning pipeline. (Twitter:@steffportelli)