design-prime-editor-sgrnas

name	design-prime-editor-sgrnas
description	Design pegRNAs (prime editing guide RNAs) and nicking sgRNAs for prime editing. Given a target gene and desired edit (SNV, small insertion, or deletion), identifies PE-compatible protospacers, designs RT templates and PBS sequences, and scores guides by predicted efficiency.
tags	["crispr","prime-editing","guide-design","pegRNA"]
version	1.0.0

Purpose

Design prime editor (PE) sgRNAs and pegRNAs for precise genomic edits without double-strand breaks. This tool automates identification of PE-compatible protospacers, design of reverse-transcriptase (RT) templates and poly(A) binding site (PBS) sequences, and scoring of predicted guide efficiency. Supports multiple PE variants (PE2, PE3, PE3b, PE-max) and custom PAM sequences.

When to Use

Use this skill when you:

• Need to perform precise edits (SNVs, small insertions, small deletions) at a known genomic locus
• Want to leverage prime editing for edits where traditional CRISPR would require DSBs
• Need to design pegRNAs with optimized PBS and RT template sequences
• Want efficiency scoring and multi-guide candidates ranked by predicted effectiveness
• Are designing for human or mouse genomes with known genomic coordinates

Do NOT use this skill when you:

• Targeting large deletions (>100bp) — use standard CRISPR-Cas9 instead
• Working with organisms without available Ensembl genome annotations (hg38, hg19, mm10)
• Need wet-lab synthesis and validation — this generates computational designs only
• Require functional validation; all predictions are computational

Expected Inputs

• Gene symbol (e.g., "TP53", "BRCA1") or explicit target sequence
• Edit specification: edit type (SNV/insertion/deletion), desired nucleotide change, position
• Genome build: hg38, hg19, or mm10
• PE architecture: PE2 (pegRNA alone), PE3 (pegRNA + nicking sgRNA)
• Optional: custom PAM sequence, PBS/RT length constraints

Expected Outputs

• TSV guide file: spacer sequence, PBS, RT template, efficiency scores, nick sgRNA (if PE3)
• Visualizations: efficiency distribution, PBS Tm distribution, genomic map of guides
• Summary: top N designs ranked by combined score

Procedure

1. Fetch genomic sequence

   • Query Ensembl REST API using gene symbol or explicit coordinates
   • Extract reference sequence spanning edit site ± 500bp

2. Identify candidate protospacers

   • Scan both DNA strands for NGG (or custom) PAM sequences
   • Filter: protospacers must be ≤30bp from edit site
   • Exclude protospacers with poly-T stretches (TTTT)

3. Design PBS and RT template

   • PBS: reverse-complement of pegRNA 3' end (8-17nt)
   • RT template: encodes desired edit + flanking context (10-40nt)
   • Calculate PBS Tm using nearest-neighbor thermodynamics

4. Score guide efficiency

   • GC content: optimal 40-70%
   • PBS Tm: optimal 30-37°C
   • Seed region (17-20bp): no off-target-prone patterns
   • Combined score: normalized average of subscores

5. Design nicking sgRNAs (PE3 only)

   • Find NGG PAMs on opposite strand
   • Distance from pegRNA: 40-100bp (optimal ~70bp)
   • Report top nicking guides alongside pegRNA

6. Generate output

   • Rank by efficiency score
   • Export TSV with all guide properties
   • Plot distributions and genomic context

Key Execution Patterns

Basic SNV design

python scripts/design_prime_editor_sgrnas.py \
  --gene TP53 \
  --species human \
  --edit-type snv \
  --position 7577121 \
  --wt-seq TCGCTAGGCGGATGCT \
  --edit-seq TCGCAAGGTGGATGCT \
  --editor PE3 \
  --top-n 20 \
  --outdir results_tp53_snv/

Small insertion with custom PAM

python scripts/design_prime_editor_sgrnas.py \
  --gene BRCA1 \
  --edit-type insertion \
  --target-sequence AGGATCAAAGAGCGAGCAGAGTC \
  --editor PE2 \
  --pam NG \
  --pbs-length-range 10,15 \
  --outdir results_brca1_insert/

Deletion using wild-type sequences

python scripts/design_prime_editor_sgrnas.py \
  --gene PTEN \
  --species mouse \
  --edit-type deletion \
  --wt-seq CACTGGTTATAAAGAGCGTAGCTAA \
  --edit-seq CACTGTAA \
  --genome mm10 \
  --editor PE3b \
  --nick-distance-range 50,80 \
  --outdir results_pten_del/

Parameter Decision Guide

Parameter	Guidance	Default	Notes
--editor	PE2: pegRNA only (low WT off-target); PE3: add nicking sgRNA (higher efficiency)	PE3	PE3b includes RT domain optimization
--pam	NGG for SpCas9 (most common); NNG for xCas9; NNGRRT for VQR	NGG	Custom PAM for alternative nuclease variants
--pbs-length-range	Longer PBS (14-17) = higher specificity; shorter (8-11) = higher activity	8,17	10-13 often best compromise
--rt-length-range	Longer RT (25-40) encodes more context; shorter (10-15) faster folding	10,40	15-25 commonly used in publications
--nick-distance-range	PE3 spacing; 60-80bp often optimal; <40bp or >100bp = reduced activity	40,100	Empirical from Anzalone et al. 2019
--spacer-len	SpCas9/PE standard is 20bp; xCas9 can use 17-20bp	20	Fixed at 20 for PE compatibility
--position	Genomic coordinate (1-indexed); required if not providing explicit sequence	—	Must match selected genome build

Failure Modes

Error	Cause	Solution
"Gene not found"	Gene symbol not in Ensembl	Verify gene name; use explicit --target-sequence instead
"No candidate guides found"	All protospacers filtered out	Relax --gc-min/--gc-max; extend search radius from edit site
"HTTP 400 from Ensembl API"	Invalid genome build or coordinates	Check --genome flag; verify --position matches build (hg38 vs hg19 differ)
PBS Tm out of range	Few guides with optimal Tm	Accept 25-40°C as acceptable range or reduce --pbs-length-range
All poly-T guides filtered	Region has high AT content	Use --target-sequence to manually specify; consider alternative edits

Implementation Details

• Sequence fetching: Ensembl REST API (no authentication required)
• PAM scanning: IUPAC regex for flexibility (e.g., "NG" = [ACGT]G)
• Tm calculation: O(N) nearest-neighbor lookup table (no external thermodynamics library)
• Scoring: Weighted combination of GC%, Tm, seed quality, position heuristics
• Plotting: matplotlib with 300 DPI PNG output

References

• Anzalone, A. V., et al. (2019). Search-and-replace genome editing. Nature 576, 149-157.
• Chen, P. J., et al. (2021). Enhanced prime editing systems by manipulating pegRNA-polymerase interactions. Nature Biotechnology 39, 1288-1293.