bio-genome-engineering-off-target-prediction

What this skill does

## Version Compatibility

Reference examples tested with: Cas-OFFinder 3.0+, pandas 2.2+, Python 3.10+.

Before using code patterns, verify installed versions match. If versions differ:
- Python: `pip show <package>` then `help(module.function)` to check signatures
- CLI: `<tool> --version` then `<tool> --help` to confirm flags

If code throws ImportError, AttributeError, or TypeError, introspect the installed package and adapt the example to match the actual API rather than retrying.

Results depend on inputs far more than tool versions: the candidate list is bounded by the **reference genome build, the mismatch/bulge tolerance, and the PAM pattern** searched, not by the Cas-OFFinder version. The **CFD matrix is SpCas9/NGG-specific and a relative ranker, not a calibrated cutting probability**. Load the published CFD tables (Doench 2016 / CRISPOR distribution) rather than hand-typing values. Cas-OFFinder is the maintained `snugel/cas-offinder` repository (native DNA/RNA bulge support from v3.0.0).

# Off-Target Prediction

**"Check my guide for off-targets"** -> Enumerate candidate sites genome-wide by mismatch/bulge tolerance, rank them by a per-site score, decide whether in-silico is sufficient or empirical discovery is required, and report each claim at the right rung: predicted, detected, or validated.
- CLI: `cas-offinder input.txt G output.txt` enumerates sites (no ranking)
- Python: CFD scoring from the published mismatch/PAM tables; aggregate specificity
- Web/CLI: `CRISPRme` for variant-aware (gnomAD + individual) nomination; `CRISPOR` to aggregate

## The Single Most Important Modern Insight -- in-silico enumeration nominates *candidates*; it does not measure *which sites are cut*

The naive model -- "search the genome within N mismatches, score by CFD, the high scorers are my off-targets" -- is wrong in three structural ways no better scoring fixes:

1. **Mismatch count is not cleavage.** A 2-mismatch site in closed chromatin may never be cut; a 3-mismatch site in open chromatin near an active promoter is. Cellular cutting depends on chromatin, dose, and exposure time -- invisible to a sequence search.
2. **Bulges and non-canonical PAMs are routinely missed.** Real validated off-targets occur with 1-2 nt DNA/RNA bulges and at NAG/NGA PAMs; fixed-alignment mismatch-only search misses them. **The failure is silent** -- a clean report looks identical whether the guide is specific or the search just couldn't see the off-target.
3. **CFD is a narrow, SpCas9-only relative ranker.** A CFD of 0.08 is not "8% chance of cutting"; comparing two guides' aggregate scores is fine, reading an absolute CFD as a safety threshold is not.

The corollary, and the central professor-level point: **in-silico lists overlap only partially with empirically validated off-targets, and the empirical genome-wide assays disagree with *each other* too.** No single method is authoritative. Off-target evidence escalates: **predicted -> detected by an unbiased assay -> validated by targeted amplicon deep-seq.** Conflating these rungs is the field's most common error. Therapeutic-grade assessment is *triangulation* (variant-aware in-silico + >=2 orthogonal empirical assays + amplicon validation + a structural readout), never one tool's output.

## In-Silico Taxonomy -- enumerate, then score, then aggregate

| Layer | Tool | Citation | Role / caveat |
|-------|------|----------|---------------|
| Enumerate | **Cas-OFFinder** | Bae 2014 *Bioinformatics* 30:1473 | exhaustive, alignment-free, GPU; **DNA/RNA bulges** (native v3.0.0); returns sites, **no ranking** |
| Enumerate (variant) | CRISPRitz | Cancellieri 2020 *Bioinformatics* 36:2001 | enumerates against genome **+ a VCF** of variants, with bulges; backend of CRISPRme |
| Enumerate (scale) | GuideScan2 | Schmidt 2025 *Genome Biol* 26:41 | genome-wide specificity databases (NOT Nat Biotechnol) |
| Score (per-site) | **CFD** | Doench 2016 *Nat Biotechnol* 34:184 | position x mismatch-type matrix x PAM penalty; **SpCas9/NGG only**, poor on bulges; de facto standard |
| Score (legacy) | MIT/Hsu | Hsu 2013 *Nat Biotechnol* 31:827 | original; **deprecated/flawed** -- report, don't lead with it |
| Score (ML) | CRISTA; Elevation | Abadi 2017; Listgarten 2018 | Elevation folds in chromatin accessibility |
| Aggregate | **CRISPOR**; **CRISPRme** | Concordet 2018 *NAR* 46:W242; Cancellieri 2023 *Nat Genet* 55:34 | CRISPOR = research one-stop; CRISPRme = variant-aware therapeutic nominator |

The unifying caveat: every score is bounded by the enumerator's coverage -- **if the enumerator didn't propose a site (bulge, distal PAM, beyond the mismatch cutoff), no scorer will ever flag it.**

## Empirical Discovery Assays -- each has a characteristic bias; concordance is partial

| Assay | Citation | Class | Bias |
|-------|----------|-------|------|
| **CIRCLE-seq** | Tsai 2017 *Nat Methods* 14:607 | in-vitro (cell-free) | **over-calls** (no chromatin); most sensitive *candidate generator* |
| **CHANGE-seq** | Lazzarotto 2020 *Nat Biotechnol* 38:1317 | in-vitro | scalable CIRCLE-seq; same over-call caveat |
| Digenome-seq | Kim 2015 *Nat Methods* 12:237 | in-vitro (WGS) | unbiased but depth-limited, expensive |
| SITE-seq | Cameron 2017 *Nat Methods* 14:600 | in-vitro | concentration series ranks sensitivity |
| **GUIDE-seq** | Tsai 2015 *Nat Biotechnol* 33:187 | cell-based (dsODN tag) | physiological; **misses rare sites**, cell-type-specific, hard in primary/RNP |
| **DISCOVER-seq** | Wienert 2019 *Science* 364:286 | cell-based (MRE11 ChIP, in situ) | tag-free, works in vivo; depends on transient MRE11 occupancy |
| TTISS | Schmid-Burgk 2020 *Mol Cell* 78:794 | cell-based | high-throughput; benchmarks fidelity variants |

**The load-bearing reality:** in-vitro assays over-call (high sensitivity, low cellular specificity); cell-based assays under-call rare sites and are cell-type-dependent (K562 yields far more hits than HEK293 for the same guide). Cross-method discordance is *information*, not noise -- sites found by both are high-confidence; in-vitro-only sites are likely chromatin-protected. The defensible workflow is the **VIVO** logic (Akcakaya 2018): sensitive in-vitro generator -> cell-based assay in the relevant cell type -> amplicon validation.

## Decision Tree by Scenario

| Scenario | Recommended | Why |
|----------|-------------|-----|
| Research knockout / screen (some off-target tolerable) | CRISPOR or GuideScan2 to pick the most specific guide; Cas-OFFinder (<=4 mm + bulges) to eyeball top sites | in-silico is sufficient when being wrong is cheap |
| Choosing among candidate guides | rank by **aggregate CFD specificity** (compare guides, not absolute safety) | specificity is a separate axis from on-target activity (-> grna-design) |
| Human therapeutic guide | **variant-aware** CRISPRme vs gnomAD (+ patient genome), bulges on | a common ancestry-enriched SNP can create a real off-target (rs114518452 / BCL11A) |
| Therapeutic, choosing the nuclease | high-fidelity variant **in the delivery format actually used** | RNP -> HiFi Cas9 (R691A) or Sniper-Cas9; plasmid-tuned variants can lose their edge as RNP |
| Therapeutic validation | >=2 orthogonal empirical assays -> amplicon deep-seq with stated LoD -> structural readout | predicted != detected != validated; amplicons miss large deletions/translocations |
| Base/prime-editor off-targets | this skill covers **Cas-dependent** only | deaminase (Cas-independent) DNA/RNA off-targets -> base-editing-design / prime-editing-design |

## High-Fidelity Nucleases -- often a bigger lever than guide reselection

| Variant | Citation | Note |
|---------|----------|------|
| eSpCas9(1.1) | Slaymaker 2016 *Science* 351:84 | neutralizes non-target-strand contacts; characterized mostly as plasmid |
| SpCas9-HF1 | Kleinstiver 2016 *Nature* 529:490 | weakens 4 Cas9-DNA H-bonds; plasmid-characterized |
| HypaCas9 | Chen 2017 *Nature* 550:407 | conformational proofreading gate |
| evoCas9