The DeepDDG server predicts the stability change of protein point mutations using neural networks.
The dataset used to train the network can be downloaded here.
For a quick demo, use the T4 lysozyme structure and the predicted result is like this.

Protein Structure (Please make sure the uploaded structure does not contain short peptides and has less than 600 residues in each chain):
Select PDB file to upload:
Or download PDB ID:

Network model:
DeepDDG (non-integrated model)
iDeepDDG (integrated model, uses user-provided ΔΔG values from mCSM, SDM, and DUET as additional inputs)

Mutation list:
All possible mutations
Selected mutations ( Multiple mutations may be submitted in one job )
Mutation list (format):
  • Each line contains one mutation, multiple lines can be submitted in one job :
    Chain WTResidueName ResidueID MutantResidueName
    e.g., A W 32 G
  • If the iDeepDDG network is used, put the ΔΔG values from mCSM, SDM, and DUET at the end of each mutation (order is important):
    e.g., A W 32 G -1.23 -2.04 -0.59
  • The mCSM, SDM, and DUET ΔΔG values can be obtained from the DUET server

Academic email (results will be sent to this email, ".com" email is not supported):

Submitting job...

H. Cao, J. Wang, L. He, Y. Qi*, and J. Z. Zhang*. (2019). DeepDDG: Predicting the Stability Change of Protein Point Mutations Using Neural Networks. J. Chem. Inf. Model.