Methylome-guided ranking + methylation-sensitive editing simulator for ThermoCas9.

ThermoCas9 — a thermophilic Cas9 from Geobacillus thermodenitrificans T12 (Roth et al., Nature 2026) — refuses to cleave when the fifth PAM cytosine is methylated, opening a route to tumor-selective editing at loci whose methylation diverges between cancer and matched normal tissue.

This site hosts the open methods paper that turns that mechanism into a probability-scale target ranking, a per-cohort atlas of candidate sites across four public methylation cohorts, a reproducibility tutorial, and an interactive cleavage simulator. Educational research; not a clinical decision-support system.

live Paper: paper-5-10j Atlas tag: memo-2026-04-22-bw Tests: 245 passing License: BSD-3

Atlas · headline

Per-cohort target shortlists

Top-100 candidate ThermoCas9 PAM sites across four public methylation cohorts, with a per-positive whole-genome rank dot-plot that visualizes the GSE69914 ESR1 reversal honestly. Searchable, filterable, downloadable.

Per-positive whole-genome rank percentiles for the three Roth Fig. 5d positives across four cohort paths.

fig4 dot-plot 4 cohorts × top-100 JSON + TSV

Methods · paper

Compositional probability-scale scoring

Plain-language walkthrough of p_targ × gap × p_trust and tie-band-aware benchmarking. Preprint PDF + repo + honest-scope block.

PDF Tag paper-5-10j

Simulator · interactive

Methylation-sensitive cleavage simulator

Interactive React-based illustration of ThermoCas9 PAM recognition, 5mCpG/5mCpC inhibition, and selective editing of hypomethylated cancer alleles in MCF-7 vs MCF-10A.

v0.4 Educational illustration

Reproduce · tutorial

Run a benchmark on a laptop

Eight-step uv run walkthrough — fresh git clone to annotated top-K shortlist in ~15 minutes. Verified against the actual CLI signatures.

~15 min No R, no Conda

Translational · roadmap

Translational hypothesis stack

Seven testable hypotheses, three concrete preclinical programs, and a Phase 0 / Phase 1 / platform-trial framework for moving methylation-sensitive Cas9 editing toward clinical use.

7 hypotheses 3 programs

Summer Project · education

Undergraduate research scopes

Bounded undergraduate and graduate project scopes — computational target discovery, in vitro cleavage assays, biomarker prediction — with a 10-week example and publication targets.

10-week scope Publication paths

How to cite

Until the bioRxiv DOI is live, cite the immutable git tag of the methods paper:

Paper (long form): Huang, A. (2026). Compositional probability-scale scoring and tie-band-aware benchmarking for methylome-guided ThermoCas9 target-site ranking. Technical memo, version paper-5-10j. github.com/AllisonH12/thermocas9
Paper (Bioinformatics-shaped): Same author/title; cite version memo-2026-04-22-bw.
Underlying biology: Roth et al., Nature 2026 — Molecular basis for methylation-sensitive editing by Cas9 (DOI 10.1038/s41586-026-10384-z).
Atlas snapshot: Cite the immutable tag memo-2026-04-22-bw for the per-cohort top-100 shortlists and the per-positive WG-rank dot-plot.