bioRxivJClub
HERACLES for fast and accurate likelihood-based reconstruction of massive lineages
bioRxiv · · Preprint · Open access
Persad, Morris
Abstract ↗AI summary
The abstract is read at the publisher; the summary is JClub's.
HERACLES reconstructs CRISPR lineage phylogenies for 150,000-cell lineages in about four hours on a single GPU, improving accuracy via hyperbolic coordinate optimization.
- Why it matters: Likelihood-based CRISPR phylogenies are accurate but computationally heavy, while distance-only methods can introduce unknown biases because distances may not reflect the editing process.
- What they did: HERACLES optimizes cell coordinates in hyperbolic space using a quartet-based composite likelihood with an explicit CRISPR editing evolutionary model, avoiding repeated full tree searches.
- The result: In simulated benchmarks it improves reconstruction accuracy, and on KP-tracer and PEtracer it yields trees consistent with anatomical annotations and temporal markers, also improving simulated DNA Typewriter lineages with sequential editing.
The findingWhy it mattersWhat they didThe result
AI summary of the abstract · may contain errors · not medical advice · read the paper.
- Open access