Title: Predicting Protein Conformational Ensembles Using AlphaFoldTemplate for Projects
Project duration: 08.2023 – 12.2023
Research Area: Life Science and Medicine
Our goal is to accurately predict the range of conformations a protein can adopt, thereby understanding its functional mechanisms. We aim to develop a method that enhances AlphaFold’s predictions, enabling the exploration of protein dynamics and the identification of transient structural states.
Proteins are dynamic molecules that can adopt multiple conformations, crucial for their function. Traditional structure prediction methods often fail to capture this conformational diversity, limiting our understanding of protein functionality and dynamics.
We applied our enhanced AlphaFold method to predict the conformational ensemble of the HER2 protein, implicated in breast cancer. Our predictions matched experimental data, demonstrating the method’s potential in drug discovery by identifying novel therapeutic targets.
We utilize AlphaFold2, a deep learning-based protein structure prediction algorithm. Our method involves manipulating the input multiple sequence alignments (MSAs) to AlphaFold, enabling it to predict a broader range of protein conformations.
This project has the potential to revolutionize our understanding of protein dynamics, opening new avenues in drug discovery, vaccine development, and the design of novel enzymes. Future research will focus on refining and applying our method to a broader range of proteins to explore the dynamic protein fold space further.