Predict enhancer activity in FASTA sequences via the Genomic Intelligence DeepSTARR API.
npx skills add https://github.com/ClawBio/ClawBio --skill gi-enhancer
Please help me install this Agent Skill. Skill: gi-enhancer Source: https://github.com/ClawBio/ClawBio/tree/main/skills/gi-enhancer Command: npx skills add https://github.com/ClawBio/ClawBio --skill gi-enhancer
Running DeepSTARR-style enhancer prediction locally requires Keras, GPU hardware, and tokenization expertise. This Skill removes that barrier by sending a FASTA sequence to the hosted Genomic Intelligence API and returning per-window enhancer activity scores in about one second, with full reproducibility artifacts.
/v1/tasks/enhancer/predict and receives per-window activity scores without local model setup.command.sh and environment.json.Ask the agent to predict enhancer activity for your FASTA file using gi-enhancer with your GI_API_KEY configured, or run the bundled Drosophila eve demo.
High-intent search queries and answers about installing and using this skill.
Run the gi_enhancer.py CLI with --input pointing to a single-record FASTA file and --output for the results directory. The skill posts the sequence to the Genomic Intelligence /v1/tasks/enhancer/predict endpoint and writes a Markdown report plus JSON result with per-window activity scores.
DeepSTARR is a deep learning model trained on Drosophila S2 cell STARR-seq data to predict enhancer activity from DNA sequence. Scores for mammalian sequences remain informative as relative rankings, but absolute values are calibrated for fly chromatin.
Yes, remote inference requires a Genomic Intelligence partner key supplied via the --api-key flag or the GI_API_KEY environment variable. A shared hackathon-tier key ships in the repo's .env.example, and individual keys can be requested from [email protected].
The API accepts sequences from 50 to 500,000 bp after whitespace stripping, enforced as a 422 validation error. The model context window is 249 bp, so shorter inputs are padded; the skill warns when input falls below that window.
No, identifiable patient data should not be submitted without an appropriate data-use agreement, since sequences are uploaded to the hosted Genomic Intelligence API. The skill is intended for research and development use, not clinical or diagnostic decisions.
BIsnake2001Scan genomic regions for transcription factor motifs using HOMER.
ammawlaQuery gnomAD allele frequencies and gene constraint scores via GraphQL API.
K-Dense-AIPredict promoters, splice sites, enhancers, chromatin state, and gene expression from DNA sequences via hosted transformer models.
NVIDIA-BioNeMoGenerate DNA sequences and extract model representations via Evo 2 NIM.