A declarative DSL and compiler for reproducible gene-editing design — with LLM agents and an MCP server so any AI agent can drive it safely.
-
Updated
Oct 12, 2026 - Python
A declarative DSL and compiler for reproducible gene-editing design — with LLM agents and an MCP server so any AI agent can drive it safely.
Conservation-constrained, escape-aware CRISPR guide design for HSV-1. Finds that the sampling resolution of the public genome corpus, not repair biology, limits how low an escape probability can be demonstrated.
Browser and command-line CRISPR-Cas9 design tool for human iPSC editing: two guides per edit, CFD-graded blocking of the repaired allele, ssODN/block/AAV donors, genotyping primers, and a hand-off file for Assured QC.
AI-assisted platform for designing CRISPR experiments — from target selection to bench-ready protocols
Computational pipeline for designing multiplexed CRISPR-Cas12a diagnostic panels targeting drug-resistant Mycobacterium tuberculosis
Motif-first CRISPR guide design workbench with genome-wide off-target search, model-based scoring, reproducible batch workflows, and desktop, web, and CLI interfaces.
Local CRISPR-Cas9 guide off-target scanner. Core PAM scan verified correct via property testing against 500 random sequences; added a bound on max_mismatches to prevent degenerate output near the random-match noise floor.
In-silico CRISPR guide design against the HIV-1 proviral genome: cross-subtype conservation, genome-wide human off-target exclusion, transcriptome similarity screening
Scalarization-free multi-objective preference alignment for generative CRISPR sgRNA design — standard DPO over Pareto-dominance preferences. Reproducible seeded pipeline, publication figures, and reviewer responses. Runs in <0.6 GB VRAM.
To associate your repository with the guide-rna topic, visit your repo's landing page and select "manage topics."