Closed-Loop Aptamer Development System
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Current aptamer development processes, such as SELEX, are time-consuming and resource-intensive, and often fail to identify the optimal aptamer for multiple targets due to the impracticality of exploring the vast space of possible aptamers, leading to low successful rates and uncertainty about in vivo functionality.
Innovation Solution
A closed-loop aptamer development system that synthesizes XNA aptamer libraries, partitions them into monoclonal compartments for target capture and sequencing, and uses machine-learning models to predict and validate aptamer sequences with high binding affinity, enabling efficient identification and validation of aptamers for specific targets.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If traditional SELEX process is used to identify aptamers, then aptamers can be obtained with high binding affinity to target molecules, but the process is extremely time-consuming and resource-intensive requiring evaluation of septillion (10^24) different aptamers
Solution Approach 1:
The patent segments the vast aptamer search space into manageable compartments, where each compartment contains a specific subset of aptamers. This compartmentalization allows parallel evaluation of multiple aptamer subsets simultaneously, dramatically reducing the time required to screen through the septillion possible aptamers while maintaining comprehensive coverage of the search space
Solution Approach 2:
The patent performs preliminary computational analysis and machine learning predictions to identify promising aptamer candidates before experimental validation. This preliminary filtering reduces the number of aptamers that require resource-intensive wet lab experiments, thereby reducing overall identification time while ensuring high-binding-affinity candidates are prioritized
2Reliability
If traditional SELEX process is used to identify aptamers, then selective binding aptamers can be obtained, but enormous resources are required to evaluate the full space of candidate aptamers
Solution Approach 1:
The patent divides the enormous aptamer library into smaller, manageable compartments that can be processed in parallel. Each compartment contains a representative subset of the full aptamer space, allowing resource-efficient evaluation while maintaining diversity. This segmentation enables the system to screen through the full septillion-space without requiring physical resources to handle all candidates simultaneously
Solution Approach 2:
The patent uses computational models and machine learning to create virtual representations of aptamer binding behavior. These computational copies allow the system to evaluate binding affinity and selectivity in silico before proceeding to physical experiments, dramatically reducing the quantity of physical aptamer material and resources required while maintaining evaluation accuracy
3Reliability
If traditional SELEX process is used, then aptamers can be identified through iterative experimental processes, but it is highly likely that optimal aptamer selection is not currently being achieved
Solution Approach 1:
The patent implements a closed-loop system where experimental results from compartment-based capture and sequencing feed back into machine learning models. These models continuously learn from the data and refine their predictions, improving aptamer selection quality over time. The feedback mechanism allows the system to identify optimal aptamers more effectively with each iteration while managing complexity through automated computational analysis
4Reliability
If the full space of septillion (10^24) different aptamers is explored, then optimal aptamer selection may be achieved, but it would take an enormous amount of resources and time
Solution Approach 1:
The patent segments the septillion aptamer space into parallelizable compartments that can be evaluated simultaneously. This segmentation transforms the sequential bottleneck into a parallel processing architecture, maintaining comprehensive coverage of the search space while dramatically improving productivity through concurrent evaluation of multiple aptamer subsets
Solution Approach 2:
The patent performs preliminary computational screening and machine learning-based prioritization to identify the most promising aptamer candidates before experimental validation. This preliminary action filters out low-probability candidates, allowing the system to focus resources on high-potential aptamers and achieve optimal selection rates without evaluating all septillion possibilities, thereby improving overall development efficiency
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
This approach significantly reduces the time and resources required for aptamer identification, enhances the chances of finding optimal aptamers with high binding affinity across multiple targets, and facilitates the validation of in silico-derived aptamers, improving the efficiency and effectiveness of aptamer development.
Implementation Method 1
separating the one or more monoclonal compartments of the compartment-based capture system that comprise the one or more targets bound to the unique aptamer from a remainder of monoclonal compartments
Data Source
AI summary
The present disclosure relates to a closed loop aptamer development system that identifies one or more aptamers observed experimentally and implements machine-learning models to identify other aptamers not observed experimentally. Particularly, aspects of the present disclosure are directed to receiving a query concerning one or more targets, acquiring a library of aptamers that potential satisfy the query, identifying a first set of aptamers from the library of aptamers that substantially or completely satisfy the query, obtaining sequence data for the first set of aptamers, generating, by a prediction model, a third set of aptamers derived from the sequence data for the first set of aptamers, validating the third set of aptamers that substantially or completely satisfy the query, and upon validating the third set of aptamers and in response to the query, providing the third set of aptamers as a result to the query.