Cas9 Protein Engineering via Machine Learning and Domain Mutations
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Current Cas9 variants, such as SpCas9 and SaCas9, face challenges in minimizing off-target effects and optimizing PAM recognition for genome editing, particularly in vivo, where packaging limitations and reduced genome coverage hinder their application in clinical gene therapy, and existing optimization methods are labor-intensive and costly.
Innovation Solution
A machine learning-guided approach is employed to engineer activity-enhanced Staphylococcus aureus Cas9 variants by mutating specific residues in the WED and PI domains, combined with in silico screens and structure-guided rational design, to enhance on-target activity and specificity, reducing the need for extensive experimental screening.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Productivity
If SpCas9 is used for genome editing, then editing efficiency is improved due to short PAM 5'-NGG-3' recognition, but off-target effects increase reducing editing accuracy
Solution Approach 1:
The patent applies parameter changes by systematically mutating specific amino acid residues in the PI and WED domains of SpCas9 to alter its binding parameters. This allows optimization of both PAM recognition efficiency and target site specificity, resolving the contradiction between high editing efficiency and low off-target effects through precise parameter adjustment of the Cas9 protein structure.
Solution Approach 2:
The patent applies local quality by focusing mutations on specific domains (PI and WED domains) rather than the entire Cas9 protein. This localized modification approach allows optimization of PAM recognition and target binding properties in specific regions while maintaining overall protein function, thereby improving both efficiency and accuracy without requiring global structural changes.
2Volume of moving object
If SpCas9 is minimized in size for in vivo applications, then packaging capacity for AAV is improved, but genome coverage and editing activity are reduced
Solution Approach 1:
The patent uses parameter changes by introducing specific point mutations in the PI and WED domains that alter the protein's functional parameters without significantly changing its overall size. This allows SpCas9 variants to maintain their packaging compatibility while improving genome coverage and editing activity through optimized amino acid sequences in critical domains.
Solution Approach 2:
The patent applies dynamics by creating variants with enhanced flexibility and adaptability in the PI and WED domains through targeted mutations. This allows the Cas9 protein to dynamically adjust its binding properties for better genome coverage while maintaining a compact size suitable for AAV packaging, resolving the contradiction between size minimization and functional performance.
3Volume of moving object
If SaCas9 is used instead of SpCas9, then protein size is reduced for better packaging, but PAM recognition is limited to longer 5'-NNGRRT-3' sequence reducing genome coverage
Solution Approach 1:
The patent applies parameter changes by mutating residues in the PI domain that directly interact with the PAM sequence. This allows SpCas9 variants to change their PAM recognition parameters from the strict 5'-NGG-3' to accommodate more diverse PAM sequences, thereby improving adaptability and genome coverage while maintaining the protein size advantage for AAV packaging.
Solution Approach 2:
The patent applies universality by engineering SpCas9 variants that can recognize multiple PAM sequence types through mutations in the PI and WED domains. This multi-functional PAM recognition capability allows a single Cas9 variant to serve multiple genome editing applications with different PAM requirements, enhancing versatility while maintaining compact size for in vivo delivery.
4Manufacturing precision
If extensive experimental screening is conducted to optimize Cas9 variants, then editing specificity is improved, but time and cost resources are significantly consumed
Solution Approach 1:
The patent applies preliminary action by using computational methods and structural analysis to pre-identify promising mutation candidates in the PI and WED domains before experimental screening. This preliminary computational filtering reduces the search space and allows focused experimental validation of a smaller number of high-priority variants, significantly reducing screening time while maintaining high editing specificity.
Solution Approach 2:
The patent applies mechanics substitution by replacing extensive wet-lab screening with computational modeling and in silico analysis. This substitution of computational methods for experimental methods allows rapid evaluation of numerous variants in silico, identifying optimal candidates with high specificity before minimal experimental validation, thereby dramatically reducing time and resource consumption.
Data Source
AI summary
The subject invention pertains to a Cas9 protein with an amino acid mutation at residues 888, 889, or a combination thereof of a WED domain and/or residues 988, 989, or a combination thereof of a PI domain. The subject invention can further pertain to a Cas9 protein with mutations at amino acid positions N986, D987, L988, L989, or any combination thereof. The subject invention also pertains to a method of enhancing the activity of KKH-SaCas9. In addition, a method of machine learning-based in silico screens for genome editing protein engineering is provided, including steps of populating a predictive machine learning model with an input dataset comprising empirical measurements of on-target activities of sgRNAs paired with a screening library of genome editing enzyme variants; running the predictive machine learning model with predefined parameters; and evaluating performance of the predictive machine learning model.