Automate ligand-aware protein sequence design with LigandMPNN.
npx skills add https://github.com/adaptyvbio/protein-design-skills --skill ligandmpnn
Please help me install this Agent Skill. Skill: ligandmpnn Source: https://github.com/adaptyvbio/protein-design-skills/tree/main/skills/ligandmpnn Command: npx skills add https://github.com/adaptyvbio/protein-design-skills --skill ligandmpnn
This skill automates ligand-aware protein sequence design, enabling researchers to tailor amino-acid sequences around bound ligands for improved binding and function.
Use the LigandMPNN workflow to generate ligand-aware designs for a given protein-ligand complex. Example commands: modal run modal_ligandmpnn.py --pdb-path protein_ligand.pdb --num-seq-per-target 16 --sampling-temp 0.1 python run.py --pdb_path protein_ligand.pdb --out_folder output/ --num_seq_per_target 16
High-intent search queries and answers about installing and using this skill.
Ligand-aware protein sequence design tailors amino acids around bound small molecules, enzyme active sites, metal coordination sites, and cofactors to optimize binding and function. It uses the LigandMPNN model to generate context-specific sequences for a given protein-ligand complex.
Running ligand-aware protein design locally requires Python 3.8 or higher, CUDA, and GPU hardware. You execute the run.py script with your protein-ligand complex PDB file to generate customized output sequences in your specified folder.
Yes, you can automate ligand-aware protein sequence design using the recommended Modal-based workflow. You execute the Modal script by providing a PDB path, the number of sequences per target, and a sampling temperature to generate tailored designs.
Optimizing protein binding pockets for cofactors and metal coordination sites requires ligand-context aware sequence design. By modeling the structural environment around these specific ligands, you can generate tailored sequences that improve binding activity and specificity.
You can specify the number of designed sequences generated per target for a protein-ligand complex using the num-seq-per-target flag. Adjusting this parameter alongside the sampling temperature controls the diversity and volume of your output sequences.
PharMolixPredict protein and protein-ligand complex structures and affinities using Boltz-2.
InternSciencePredict protein-ligand binding affinity from a protein sequence and SMILES using Boltz-2.
junior1pGenerate ligand-aware protein sequences for defined binding pockets from PDB files.
zongtingweiDesigns ligand-aware protein sequences for structure-guided binding pocket optimization.
