Generate amino acid sequences that fold into a specified backbone using ProteinMPNN.
npx skills add https://github.com/yjcyxky/biominer-skills --skill proteinmpnn-yjcyxky
Please help me install this Agent Skill. Skill: proteinmpnn Source: https://github.com/yjcyxky/biominer-skills/tree/main/proteinmpnn Command: npx skills add https://github.com/yjcyxky/biominer-skills --skill proteinmpnn-yjcyxky
ProteinMPNN enables designers to generate amino acid sequences that fold into a specified backbone, accelerating structure-based design workflows.
Run a design job using the provided Docker image ghcr.io/open-prophetdb/proteinmpnn:arm64-blackwell, mapping your backbone.pdb to output sequences under /data/output, for example: docker run --rm --gpus all -v $(pwd):/data ghcr.io/open-prophetdb/proteinmpnn:arm64-blackwell python /app/protein_mpnn_run.py --pdb_path /data/backbone.pdb --out_folder /data/output --num_seq_per_target 10
High-intent search queries and answers about installing and using this skill.
To design protein sequences for a specific backbone, provide a PDB file to a structure-based design tool. This Skill generates amino acid sequences that fold into your specified target, accelerating de novo protein design workflows.
Yes, multi-chain protein design is fully supported. You can perform complex tasks like interface engineering and homooligomer design while constraining fixed positions and applying amino acid biases during sequence generation.
Yes, a GPU is required for structure-based protein sequence generation. The process relies on a Docker-based image and GPU acceleration to execute model inference and generate output design files efficiently.
When designing protein sequences from a backbone PDB, the tool generates FASTA files and scoring data. These output files are saved to your specified directory for downstream sequence screening and analysis.
To run ProteinMPNN using Docker for protein engineering tasks, execute the provided Docker image command. Map your local backbone PDB file to the container and specify an output folder to generate sequences.
NVIDIA-TAOExecute TAO SDK jobs in Docker containers with NVIDIA GPU acceleration.
mims-harvardDesign protein therapeutics with RFdiffusion, ProteinMPNN, and ESMFold validation.
CASIA-LMAnalyze protein structures and functions using PDB database tools.
wuyoscarAutomate toxin protein design and scoring with PyRosetta pipelines.
