An open-source scoring library to catch unintended edits before the lab bench.
Overview
Off-target CRISPR edits are still hard to predict cheaply. This toolkit aims to give small labs a free, decent-accuracy first pass before committing to expensive validation.
Objectives
- Survey existing off-target scoring approaches
- Build an open, minimal-dependency scoring library
- Benchmark against published off-target datasets
Research questions
How much accuracy do small labs actually lose by using a lightweight open scorer instead of proprietary tools?
Methodology
Literature review followed by benchmarking existing open algorithms against public off-target validation datasets.
Current progress
Literature review nearly complete; algorithm selection underway.
Future work
Implement, benchmark, and release v0.1 of the scoring library.