메뉴
Search and select enzyme sequence homologs from a database. Use when users provide a query sequence and want to discover sequence homologs of the query sequence in a database.
Codex 또는 Claude로 설치 이 Prompt를 복사해 Codex, Claude 또는 다른 어시스턴트에 붙여 넣으면 Skill 페이지를 검토하고 설치를 진행할 수 있습니다.
SOC 직업 분류 기준
Scientific reasoning via Jnana CoScientist — hypothesis generation, evaluation, and parameter bounding
Molecular dynamics simulations using OpenMM with MDAgent-style automation and free energy calculations via MM-PBSA
Convert scientific papers into executable computational workflows and MCP tools
Protein language model embeddings, diversity sampling, and mutation prediction using ESM or GenSLM
Protein structure prediction using Chai-1 or AlphaFold with confidence scoring, quality assessment, and critic evaluation
Analysis of molecular dynamics trajectories (RMSD, RMSF, contacts, hotspot analysis)
| name | discovering-enzymes |
| description | Search and select enzyme sequence homologs from a database. Use when users provide a query sequence and want to discover sequence homologs of the query sequence in a database. |
Enzyme discovery toolkit to search, align, filter, and select enzyme sequence homologs in a database.
query.fasta) should have at least one query sequence in FASTA format.targetDB) should be in the MMseqs2 database format. A database can be downloaded by running mmseqs databases <name> <o:sequenceDB> <tmpDir>. For example: `mmseqs databases UniProtKB databases/UniProtKB tmpStep 1: Search database
Searches a local database for sequence homologs of the query in the input file and saves the homologs in FASTA format. Requires MMSeqs2 (mmseqs).
python skills/discovering-enzymes/scripts/search_database.py query.fasta \
--db UniProtKB \
--targetDB databases/UniProtKB \
--iterations 1 \
--outdir homologs/ \
Step 2: Select homologs
Generates sequence alignments, filters the alignment and selects a number of sequence homologs. Saves the selected homologs in FASTA format. Requires MMSeqs2 (mmseqs) and HHfilter (hhfilter)
python skills/discovering-enzymes/scripts/select_homologs.py query.fasta homologs.fasta \
--hhfilter software/hhfilter \
--cov 50 \
--n 48 \
--outdir homologs/ \
db: Name of a database - UniProtKB, NR, or swissprot (default: UniProtKB)targetDB: Path to the database in the MMseqs2 database format (default: databases/UniProtKB)iterations: Number of iterations for a search (default: 1)outdir: Output directory (default: homologs/)hhfilter: Path to the HHfilter executable (default: software/hhfilter)cov: 'Minimum coverage with the query sequence (default: 50)n: 'Number of sequence homologs to be selected (default: 48)outdir: Output directory (default: homologs/)homologs/
├── homologs.fasta # All homologs from a search
└── homologs_DB_filtered_selected.fasta # N selected homologs after filtering
biopython
mmseqs): Many-against-Many sequence searching and alignmenthhfilter): Filter an alignment by maximum sequence identity and minimum coverage