Design ligand-binding proteins with LigandMPNN and output PDB files.
npx skills add https://github.com/PKU-YuanGroup/OpenAI4S --skill ligandmpnn-pku-yuangroup
Please help me install this Agent Skill. Skill: ligandmpnn Source: https://github.com/PKU-YuanGroup/OpenAI4S/tree/main/skills/ligandmpnn Command: npx skills add https://github.com/PKU-YuanGroup/OpenAI4S --skill ligandmpnn-pku-yuangroup
The ligandmpnn Skill addresses the challenge of designing proteins with specific ligand-binding sites, enabling scientists to tailor protein structures to accommodate small molecules or cofactors.
Run the ligandmpnn skill with the following command: ligandmpnn run.py --model_type ligand_mpnn --checkpoint_ligand_mpnn ./model_params/ligandmpnn_v_32_010_25.pt --pdb_path complex.pdb --out_folder out --batch_size 8 --number_of_batches 4 --temperature 0.1 --fixed_residues "A45 A46 A47 A48"
High-intent search queries and answers about installing and using this skill.
You can design a ligand-binding protein by using LigandMPNN to redesign residues around a bound ligand or cofactor within a PDB complex. This neural network tailors protein structures to accommodate specific small molecules.
The best way to design metal-coordinating sites is using a neural network that incorporates metal context, such as LigandMPNN. It enables you to specify geometry and redesign protein structures for therapeutic applications requiring metal binding.
Yes, you need PyTorch and Biopython installed, along with numpy, ProDy, ml_collections, and dm-tree. These dependencies are required to execute the neural network scripts and process the structural biology data.
Yes, you can fix specific residues when redesigning a protein binding pocket by using the fixed_residues parameter. For example, you can specify residues like "A45 A46" to keep them static while the network redesigns the surrounding context.
Incorporating nucleic-acid context improves biomolecular engineering by allowing the neural network to account for structural interactions with DNA or RNA. This ensures the designed protein binding pockets are structurally compatible with the surrounding molecular environment.
The input complex should be in PDB format, typically provided via the pdb_path argument. The skill processes this structural file to redesign the sequence and generates threaded designed-sequence PDB files as output.
adaptyvbioAutomate ligand-aware protein sequence design with LigandMPNN.
yf8578Parse FASTA, GenBank, and PDB files for molecular biology analyses.
galeepSimulate and analyze molecular dynamics with OpenMM and MDAnalysis.
zongtingweiDesigns ligand-aware protein sequences for structure-guided binding pocket optimization.
