Gene Fusion Candidate Filtering for Faster Machine Learning Validation
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing methods for detecting gene fusions are computationally intensive, requiring extensive processing resources and time, which hinders efficient identification and analysis of valid gene fusions for diagnostic and therapeutic purposes.
Innovation Solution
A system utilizing a filtering engine to reduce the number of gene fusion candidates, combined with a hardware-accelerated read alignment unit and machine learning model, to quickly identify valid gene fusions, thereby reducing computational resources and runtime.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If existing methods for detecting gene fusions are used, then detection accuracy can be maintained, but computational resources and processing time become excessive
Solution Approach 1:
The detection process is divided into multiple filtering stages. A first filter reduces the initial set of fusion candidates to a smaller subset, which is then processed by a machine learning model. This segmentation of the processing pipeline allows for efficient computation while maintaining accuracy by processing only the most promising candidates through computationally intensive analysis.
Solution Approach 2:
The patent extracts and removes obviously incorrect fusion candidates through filtering mechanisms before they reach the machine learning model. By taking out false positives early in the process, the system reduces the computational burden on subsequent processing stages while maintaining high detection accuracy for valid fusions.
2Measurement precision
If existing methods for detecting gene fusions are used, then detection accuracy can be maintained, but computational resources and power consumption become excessive
Solution Approach 1:
The computational process is segmented into efficient filtering operations followed by targeted machine learning analysis. This segmentation ensures that heavy computational resources are allocated only to the subset of fusion candidates that pass initial filters, significantly reducing overall energy consumption while maintaining detection accuracy.
Solution Approach 2:
The system applies partial filtering action by using computationally lightweight filters to pre-screen candidates before applying more resource-intensive machine learning models. This partial action approach reduces total computational resource usage by avoiding exhaustive analysis of all candidates.
3Productivity
If a filtering engine is applied to reduce fusion candidates, then processing resources and runtime are reduced, but the complexity of the system increases
Solution Approach 1:
A filtering engine is introduced as an intermediary component between the raw fusion candidate generation and the machine learning model. This intermediary layer simplifies the overall system architecture by handling preliminary screening operations, making the system more modular and easier to manage despite the added component.
Data Source
AI summary
Methods, systems, and apparatuses, including computer programs for identifying a gene fusion in a biological sample are disclosed. The method can include actions of obtaining first data that represents a plurality of aligned reads, identifying a plurality of fusion candidates included within the obtained first data, filtering the plurality of fusion candidates to determine a filtered set of fusion candidates, for each particular fusion candidate of the filtered set of fusion candidates: generating, by one or more computers, input data for input to a machine learning model that includes extracted feature data that to represents the particular fusion candidate, providing the generated input data as an input to the machine learning model that has been trained to generate output data representing a likelihood that a fusion candidate is a valid gene fusion, and determining whether the particular fusion candidate corresponds to a valid gene fusion based on the output data.