Predicts gene and transcript structures from DNA sequences via the Genomic Intelligence annotation API.
npx skills add https://github.com/ClawBio/ClawBio --skill gi-annotation
Please help me install this Agent Skill. Skill: gi-annotation Source: https://github.com/ClawBio/ClawBio/tree/main/skills/gi-annotation Command: npx skills add https://github.com/ClawBio/ClawBio --skill gi-annotation
Predicting gene boundaries, exons, and transcripts from raw DNA sequence normally requires installing tools like AUGUSTUS or Helixer with species-specific models. This skill submits a FASTA region to the hosted Genomic Intelligence annotation pipeline and returns predicted transcript structures in about 20 seconds for a 20 kbp region.
Ask the agent to annotate the bundled TP53 demo sequence with gi-annotation, or provide your own single-record FASTA file between 1,000 and 500,000 bp along with a GI_API_KEY.
High-intent search queries and answers about installing and using this skill.
Provide a single-record FASTA file of 1,000 to 500,000 bp and run the gi-annotation CLI with --input and --output flags. The skill submits the sequence to the Genomic Intelligence API asynchronously and returns predicted transcripts with start, end, and strand in a Markdown report.
The skill requires a Genomic Intelligence partner key set in the GI_API_KEY environment variable or passed via the --api-key flag. A shared hackathon key ships in .env.example, and individual keys can be requested at [email protected].
The API accepts sequences from 1,000 to 500,000 bp, counted after whitespace is stripped. Out-of-range inputs return a 422 validation_failed error, and the skill rejects them locally before sending a request.
No, the underlying model is trained on human and a few other vertebrates, so bacterial, fungal, and plant predictions are out of distribution. For those organisms, use species-specific tools like AUGUSTUS with appropriate models.
The annotation pipeline is the heaviest Genomic Intelligence model, so the first request after a cold service takes over 30 seconds. Subsequent calls are warm and typically finish in about 20 seconds for a 20 kbp region.
No, it is intended for research and development use only, not clinical or diagnostic decisions. Predicted gene structures are model outputs, so clinical interpretation should anchor to curated RefSeq or Ensembl annotations.
SologaRead, write, and manipulate genomic data files in SAM, BAM, CRAM, VCF, BCF, FASTA, and FASTQ formats.
K-Dense-AIPredict promoters, splice sites, enhancers, chromatin state, and gene expression from DNA sequences via hosted transformer models.
jaechang-hitsRetrieve genomic and transcript annotations from the Ensembl REST API.
qinyan-aiQuery Ensembl REST API for gene, sequence, and variant data across species.
