Engineered ADAR Guide RNA Design for Specific RNA Editing

Resolve Bottlenecks,
Find Innovative SolutionsGenerate Solutions

Solution Overview

Problem

Current RNA editing systems using adenosine deaminase acting on RNA (ADAR) enzymes face limitations such as aberrant effector activity, delivery barriers, unintended transcriptomic modifications, and immunogenicity, necessitating improved efficiency, specificity, and safety in targeted RNA editing.

Innovation Solution

A machine learning-based approach is employed to design engineered guide RNAs (gRNAs) that predict on-target editing efficiency and specificity by utilizing a sequence and structural features, personalized to patient-specific sequences or common disease-causing mutations, using a pipeline that integrates supervised learning with high-throughput screening to optimize gRNA designs.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Reliability

If current RNA editing systems using ADAR enzymes are used, then RNA editing function is achieved, but aberrant effector activity and unintended transcriptomic modifications occur

Engineering Contradiction:

Improveediting specificityVSAvoidaberrant effector activity

Core Design Contradiction:

ReliabilityVSObject-generated harmful factors

Solution Approach 1:

The patent applies local quality by designing gRNAs with specific structural features (hairpin structures, stem-loop configurations) that concentrate ADAR enzyme activity at the target site while preventing off-target effects. The gRNA structure is optimized locally around the target binding region to enhance specificity and reduce aberrant editing activity in non-target regions.

Inventive Principle:

Principle #3Local quality

Solution Approach 2:

The patent employs parameter changes by systematically optimizing gRNA sequence parameters (length, GC content, secondary structure elements) and structural parameters to modulate ADAR enzyme recruitment and editing specificity. Machine learning models are used to predict and optimize these parameters for maximizing on-target editing while minimizing off-target effects.

Inventive Principle:

Principle #35Parameter changes

2Reliability

If current RNA editing systems are used, then editing function is achieved, but delivery barriers and immunogenicity issues arise

Engineering Contradiction:

Improveediting safetyVSAvoidimmunogenicity

Core Design Contradiction:

ReliabilityVSObject-affected harmful factors

Solution Approach 1:

The patent uses copying by designing synthetic gRNA molecules that replicate the structural and functional features of endogenous RNA structures recognized by ADAR enzymes. These engineered gRNAs mimic natural RNA secondary structures (hairpins, stem-loops) to evade immune detection while maintaining editing functionality, thereby reducing immunogenicity.

Inventive Principle:

Principle #26Copying

3Measurement precision

If machine learning-based gRNA design is implemented, then predictive accuracy of gRNA performance is enhanced, but computational complexity and model training requirements increase

Engineering Contradiction:

ImprovegRNA performance prediction accuracyVSAvoidcomputational system complexity

Core Design Contradiction:

Measurement precisionVSDevice complexity

Solution Approach 1:

The patent applies preliminary action by pre-training machine learning models on large datasets of gRNA sequences and their corresponding editing outcomes before actual gRNA design. The models are pre-trained to recognize patterns and features that correlate with successful editing, enabling rapid and accurate prediction of gRNA performance without requiring complex real-time computations during the design phase.

Inventive Principle:

Principle #10Preliminary action

Applied Scientific Principles

This section explains which scientific principles are used to turn an abstract innovation direction into a practical engineering solution.

Function Achieved in This Case

The method enhances the predictive accuracy of gRNA performance, enabling targeted ADAR-mediated RNA editing with improved efficiency and specificity, shortening discovery timelines for therapeutic applications.

Implementation Method 1

a machine learning model... receives various inputs such as a sequence of a gRNA and a sequence of the target RNA... outputs a predicted percentage of on-target editing of a desired nucleotide and a predicted specificity score

Methodology Applied

Scientific EffectMachine learning prediction:

Implementation Method 2

an engineered guide RNA (gRNA) comprising a sequence that has a predicted percentage of on-target editing of a desired nucleotide... receives various inputs such as a sequence of a gRNA and a sequence of the target RNA

Methodology Applied

Scientific EffectRNA hybridization:

Implementation Method 3

RNA editing is a post-transcriptional process that recodes hereditary information by changing the nucleotide sequence of RNA molecules... One form of post-transcriptional RNA modification is the conversion of adenosine-to-inosine (A-to-I), mediated by adenosine deaminase acting on RNA (ADAR) enzymes

Methodology Applied

Scientific EffectAdenosine-to-inosine editing: Enzyme

Data Source

PatentUS20250356949A1Machine-learning based design of engineered guide systems for adenosine deaminase acting on RNA editing

Publication Date: 2025.11.20 SHAPE THERAPEUTICS INC

AI summary

Systems and methods for predicting deamination efficiency or specificity associated with a guide RNA (gRNA) are provided. A nucleic acid sequence for the gRNA is received. Responsive to inputting a data structure into a model, a metric for an efficiency or specificity of deamination by a first Adenosine Deaminase Acting on RNA (ADAR) protein of a target nucleotide position in mRNA transcribed from a target gene is obtained as output from the model. The data structure includes an encoding of the nucleic acid sequence for the gRNA.