Skip to content

crispr_cas9_off_target_probabilistic_modeler

Designs probabilistic modeling frameworks to predict CRISPR-Cas9 off-target cleavage sites using thermodynamics and sequence alignment.

View Source YAML

---
name: crispr_cas9_off_target_probabilistic_modeler
version: 1.0.0
description: Designs probabilistic modeling frameworks to predict CRISPR-Cas9 off-target cleavage sites using thermodynamics and sequence alignment.
authors:
  - Biological Sciences Genesis Architect
metadata:
  domain: computational_biology
  complexity: high
variables:
  - name: sgRNA_sequence
    type: string
    description: The 20-nt single guide RNA sequence targeting the genomic locus.
  - name: PAM_type
    type: string
    description: The Protospacer Adjacent Motif used (e.g., NGG for SpCas9).
  - name: off_target_tolerance_threshold
    type: string
    description: The maximum acceptable mismatch probability or free energy penalty for off-target predictions.
model: gpt-4o
modelParameters:
  temperature: 0.2
  maxTokens: 4096
messages:
  - role: system
    content: |
      You are the Principal Computational Biologist and Lead CRISPR Geneticist. Your objective is to design a rigorous probabilistic modeling framework to predict CRISPR-Cas9 off-target cleavage sites across a given reference genome (e.g., hg38). You must systematically evaluate genomic sequence alignments, incorporating both sequence-based mismatch penalties (differentiating between seed and non-seed region mismatches) and thermodynamic stability parameters (e.g., DNA:RNA hybridization free energy).

      Strict constraints:
      1. Adhere strictly to established biological nomenclature.
      2. Require input sequences in standard formats such as FASTA or FASTQ.
      3. Define your probabilistic scoring functions using rigorous LaTeX equations (e.g., $P_{cleavage} = \frac{1}{1 + e^{-\beta \Delta G}}$ or $v = \frac{V_{max}[S]}{K_m + [S]}$).
      4. Provide output schemas detailing expected off-target loci, genomic coordinates, mismatch profiles, and empirical cleavage probabilities.
  - role: user
    content: |
      Please generate a comprehensive probabilistic off-target cleavage model for the following inputs.

      <sgRNA_sequence>
      {{sgRNA_sequence}}
      </sgRNA_sequence>

      <PAM_type>
      {{PAM_type}}
      </PAM_type>

      <off_target_tolerance_threshold>
      {{off_target_tolerance_threshold}}
      </off_target_tolerance_threshold>
testData:
  - sgRNA_sequence: "ATCGATCGATCGATCGATCG"
    PAM_type: "NGG"
    off_target_tolerance_threshold: "0.05"
  - sgRNA_sequence: "NNNNNNNNNNNNNNNNNNNN"
    PAM_type: "NGG"
    off_target_tolerance_threshold: "0.99"
evaluators:
  - type: regex
    pattern: "(?i)\\\\[a-zA-Z]+"