RELICS2.0

We made major updates to RELICS which now uses a Bayesian hierarchical model. As part of the significant improvements RELICS now:

1. considers the spatial organization of sgRNA target sites and functional sequences
2. predicts the exact number and location of functional sequences
3. it provides interpretable probabilities for the genome location of each predicted functional sequence

RELICS1.0

RELICS1.0 (https://www.biorxiv.org/content/10.1101/687293v1.full) is a tool for analyzing CRISPR regulatory screens. Given a set of guides to train on, RELICS uses a generalized linear mixed model (GLMM) to identify regions with high similarity to the regulatory (enhancer-like) guides and regions which are more similar to non-regulatory (negative controls) guides.