Protein Variant Generation via Allosteric Microswitch Scoring

Resolve Bottlenecks,
Find Innovative Solutions
Generate Solutions

Solution Overview

Problem

Existing methodologies for predicting and engineering protein variants lack predictive power and rationale, often destabilizing the resting state or failing to exhibit ligand-induced signal transduction, leading to inefficient screening and biased search for new drugs.

Innovation Solution

A computer-implemented method that generates protein variants by identifying allosteric sites (microswitches) and computing scores for allosteric coupling and stability, predicting activity changes, and testing in vitro to optimize selectivity, specificity, and sensitivity.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Stability of the object's composition

If empirical screening approaches are used to identify stabilized mutants, then protein stability is improved, but ligand-induced signal transduction response is lost

Engineering Contradiction:
Improveprotein stabilityVSAvoidsignal transduction response
Core Design Contradiction:
Stability of the object's compositionVSReliability

Solution Approach 1:

The method performs preliminary computational analysis to identify allosteric sites and predict mutation effects before actual mutagenesis is performed. By using in silico modeling to pre-screen potential mutations and predict their impact on both stability and signaling, the approach avoids random screening and targets mutations that are likely to maintain both stability and functional response.

Inventive Principle:
Principle #10Preliminary action

Solution Approach 2:

The method introduces computational modeling and prediction algorithms as intermediaries between the goal of stabilizing mutants and the requirement for maintaining signal transduction. These computational tools serve as mediators that can predict whether a mutation will affect stability or signaling before the mutation is actually introduced, allowing for informed selection of mutations that satisfy both criteria.

Inventive Principle:
Principle #24Intermediary (Mediator)

2Reliability

If scanning mutagenesis is performed to identify constitutively activating mutations, then receptor activity is improved, but predictive power is lost

Engineering Contradiction:
Improvereceptor activityVSAvoidpredictive power
Core Design Contradiction:
ReliabilityVSLoss of information

Solution Approach 1:

The methodology performs preliminary computational identification of allosteric sites and prediction of mutation effects before conducting mutagenesis experiments. By using in silico approaches to pre-identify promising mutation targets based on their predicted impact on receptor activity, the method maintains predictive power while systematically improving receptor function.

Inventive Principle:
Principle #10Preliminary action

Solution Approach 2:

The method replaces random mechanical screening approaches with computational prediction systems. Instead of relying on brute-force screening of numerous mutations, the approach uses computational models to predict which mutations will achieve desired activity levels, thereby maintaining and enhancing predictive power while improving receptor activity.

Inventive Principle:
Principle #28Mechanics substitution (Replace mechanical system)

3Reliability

If random mutations are generated to stabilize active state, then constitutive activity is improved, but structural coupling is destabilized

Engineering Contradiction:
Improveconstitutive activityVSAvoidstructural coupling
Core Design Contradiction:
ReliabilityVSStability of the object's composition

Solution Approach 1:

The method performs preliminary computational analysis to identify allosteric sites and predict how mutations will affect both constitutive activity and structural coupling. By using in silico modeling to evaluate potential mutations before implementation, the approach can select mutations that enhance constitutive activity while preserving necessary structural couplings.

Inventive Principle:
Principle #10Preliminary action

Solution Approach 2:

Computational prediction tools serve as intermediaries that evaluate the dual impact of mutations on both constitutive activity and structural coupling. These intermediaries provide predictive information about how a mutation will affect multiple properties simultaneously, enabling selection of mutations that achieve desired activity while maintaining structural integrity.

Inventive Principle:
Principle #24Intermediary (Mediator)

Data Source

PatentEP4004925B1Method for generating variants of a protein
Publication Date: 2025.06.25 ECOLE POLYTECHNIQUE FEDERALE DE LAUSANNE (EPFL)
  • EP4004925B1 patent drawingFigure 1
  • EP4004925B1 patent drawingFigure 2
  • EP4004925B1 patent drawingFigure 3

AI summary

The present relates to a method for generating variants of a protein based on a native protein regulated by allosteric pathway, the method comprising: - i) providing 3D structures of the native protein; - ii) identifying at least one pair of coupled allosteric sites within the amino acid sequence of the native protein named microswitch; - iii) generating in silico mutations of said identified microswitch to generate a pool of variants; - iv) computing at least one score reflecting the variation in allosteric coupling; and/or the variation in the relative stability - v) predicting the activity of each variant compared to the native protein based on the computed score. The invention also concern a computer implemented program to carry out said method, and a variant of a protein or an active fragment thereof, a polynucleotide.