
by Baker Lab
LigandMPNN provides inference code for the LigandMPNN and ProteinMPNN models, which design protein sequences from a structure. Code and parameters are MIT licensed.
Web appFree
Step 1: Clone the repository and run the script that downloads the model parameters.
Step 2: Set up an environment and install the requirements.
Step 3: Run run.py with a seed, an input PDB path and an output folder.
Step 4: Choose a model type and options such as temperature or fixed residues.
Step 5: Read the designed sequences and confidence values from the output files.
LigandMPNN, ProteinMPNN, a soluble protein model, two membrane-aware variants and a side chain packing model.
Yes. You can fix, bias or redesign residues using chain and residue index, such as A23.
They reflect the model's average probability over the redesigned residues, from 0 to 1, where higher means more confident.
The default settings run ProteinMPNN.
Boltz-1Boltz builds frontier AI models for biomolecular structure prediction and design, offered through Boltz Lab and the Boltz API. You can design small molecules and proteins and predict structures.
DeepChemDeepChem is an open-source Python library for machine learning and deep learning on molecular and quantum datasets. It builds on PyTorch and other popular frameworks.
MatterGenMatterGen is a generative model from Microsoft for designing inorganic materials across the periodic table. It can be fine-tuned to steer generation toward property constraints.