Predict protein and nucleic acid structures using the ESMFold2 algorithm.
npx skills add https://github.com/PKU-YuanGroup/OpenAI4S --skill esmfold2-pku-yuangroup
Please help me install this Agent Skill. Skill: esmfold2 Source: https://github.com/PKU-YuanGroup/OpenAI4S/tree/main/skills/esmfold2 Command: npx skills add https://github.com/PKU-YuanGroup/OpenAI4S --skill esmfold2-pku-yuangroup
This Skill provides a streamlined workflow for predicting protein and nucleic acid structures using the ESMFold2 algorithm, simplifying the process of structural biology research.
Run the 'esmfold2' skill with the following command to predict the structure of a protein with a given sequence: esmfold2 predict protein "sequence" --name "output.cif" --model "biohub/ESMFold2".
High-intent search queries and answers about installing and using this skill.
To predict protein structure from a single sequence, use the ESMFold2 co-folding method by running the predict command with your sequence and specifying an output name. This provides potential structural configurations directly from the sequence input.
Yes, you can predict nucleic acid structures including DNA and RNA using ESMFold2. The skill supports multiple input modes for protein, DNA, RNA, and ligand analysis, utilizing advanced language models for structural biology research.
Yes, you need a GPU to run ESMFold2 for structure prediction because the algorithm requires GPU acceleration. You also need specific Python and CUDA versions installed to execute the prediction workflows properly.
Single-sequence mode predicts structures using only one input sequence, while MSA mode uses multiple sequence alignment for analysis. ESMFold2 supports both input modes to accommodate different protein, DNA, RNA, and ligand analysis needs.
ESMFold2 analyzes protein mutations and contacts by incorporating ESMC language models to compute masked-LM logits, hidden states, mutation scoring, contact prediction, and SAE interpretability head for detailed sequence analysis.
You should use ESMFold2 in your drug discovery workflow when you need to predict protein and nucleic acid structures. It is suitable for structural biology research and streamlines the process of obtaining structural configurations.
junior1pPredict 3D structures for protein, DNA, RNA, ligand, and ion complexes.
yf8578Query the AlphaFold EBI API for protein structures and pLDDT scores by UniProt ID.
adaptyvbioPredict protein structures from FASTA sequences using Chai-1.
hxk622Generate protein sequences and structures with ESM3 and embeddings with ESM C.
