Accelerate Protein Stability Engineering with ThermoMPNN on Vecura
This update empowers protein engineers and computational biologists to conduct high-throughput, in silico site-saturation mutagenesis through a streamlined workflow on Vecura, eliminating the need to manage complex computational pipelines or heavy infrastructure.
What is ThermoMPNN?
ThermoMPNN is a graph neural network designed to predict thermodynamic stability changes (ddG) for every possible single-point mutation in a protein. By leveraging transfer learning on ProteinMPNN structure embeddings and training on the large-scale Megascale dataset, it enables rapid site-saturation mutagenesis in silico.
It helps users understand the effect of amino acid substitutions on protein stability without the need for high-throughput wet-lab experiments. It is especially useful for protein engineers and researchers looking to identify stabilizing or destabilizing mutations across an entire protein sequence efficiently.
What can users do with ThermoMPNN on Vecura?
With ThermoMPNN on Vecura, users can:
- Perform a comprehensive site-saturation mutagenesis (SSM) scan by inputting a protein structure in PDB format.
- Generate a complete ddG landscape for a specific chain to identify candidates for stabilizing mutations.
- Effortlessly rank all 20 possible amino acid substitutions at every residue position.
- Prioritize the most impactful mutations for downstream experimental validation through an intuitive, streamlined workflow.
What the output means
The output provides a CSV file containing site-saturation mutagenesis predictions. Each entry lists the position, wild-type residue, mutant residue, and the predicted change in folding free energy (ddG) in kcal/mol. Negative values indicate predicted stabilization, while positive values suggest destabilization.
This output should be used to support scientific decision-making. It does not replace experimental validation.
Why this matters
The ability to accurately predict the impact of point mutations on protein stability is a cornerstone of protein engineering. Traditional experimental methods to map these effects are time-consuming and expensive. ThermoMPNN significantly accelerates this process by providing reliable, high-throughput computational screenings, allowing researchers to explore vast mutational spaces with minimal resource expenditure.
By integrating this state-of-the-art model into Vecura, we remove the technical barriers associated with complex model deployment, empowering scientists to focus on therapeutic design, enzyme optimization, and fundamental protein research.
- Developed by: The Kuhlman Lab
- Source: Official GitHub Repo
- Reference: Dieckhaus, M., et al. (2024). "ThermoMPNN: a graph neural network for the prediction of thermodynamic stability changes." PNAS. https://doi.org/10.1073/pnas.2314853121
Try ThermoMPNN on Vecura.
Open the model workspace and start evaluating it with your own inputs.