Novel AI approach to unravel complex protein folding landscapes: towards better bioeconomy
EMSL Project ID
60288
Abstract
Many proteins, especially enzymes, need to bind metal ion cofactors, or gain other posttranslational modifications (PTM), to be active. None of these processes are static; protein structures are dynamic and reversible PTMs can be used for cell signaling and/or destroy protein function. Even if we have a lot of data today on protein structures, and AlphaFold2 has revolutionized the predictive ability, there are many unknowns around protein conformational dynamics and its relation to PTMs. To address this, here we will create a full atomistic folding energy landscape for a copper-binding redox protein using an AI-ML approach informed by experimental data. The target is the bacterial protein azurin, for which extensive in vitro biophysical data exist and thus it is an excellent starting point, but the overall goal is to extend this method to catalytic metalloproteins used in biomass degradation and in cell factories to produce biofuels and fine chemicals. In addition to metal modulation, other post-translational modifications, such as cysteine oxidation that often occurs in microbial systems, on folding landscapes will be explored (azurin has three redox-active cysteines). The unique concept we will explore involves the input of experimental kinetic data on folding of apo and metal-loaded azurin in the form of phi-values, that report on folding-transition state structures, into a new AI/ML code created by a team member at ANL. With the aid of EMSL expertise, the code will be used together with the experimental data to reveal atomistic details of the folding dynamics of azurin as a function of PMTs. The work will then be extended to a related copper-binding redox-active protein, LPMO (lytic polysaccharide monooxygenase), that is currently used in industrial biomass degradation processes. Notably, many enzymes used in bioeconomy and bioenergy processes are not working at full capacity and improvement is desired. Thus, this project is of high environmental relevance and may help improve enzymes used in bioeconomy processes.
Project Details
Project type
Large-Scale EMSL Research
Start Date
2022-10-01
End Date
N/A
Status
Active
Released Data Link
Team
Principal Investigator
Team Members