In Silico circRNA Junction Validation via Pseudo-Reference Alignment
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Current computational algorithms for detecting circular RNAs (circRNAs) face challenges due to divergent results from different bioinformatics methods and sequencing data treatments, necessitating novel methods for in silico validation and selection of circRNA candidates for further analysis.
Innovation Solution
A bioinformatics method involving the reception of RNA sequencing data, extraction of start-reads and end-reads, assembly into contigs, generation of a pseudo-reference, and alignment to validate circRNA junctions, using tools like Trinity and BWA-MEM, to identify overlaps and select circRNAs for further analysis or experimental validation.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If current computational algorithms are used for detecting circRNAs, then circRNA candidates can be identified, but the results diverge across different bioinformatics methods and sequencing data treatments
Solution Approach 1:
The patent introduces an intermediary validation process using extracted reads and contig assembly as a mediator between initial circRNA detection and final validation. This intermediary step involves assembling reads into contigs and comparing them against a reference genome to verify circRNA junctions, thereby resolving the divergence between different detection methods.
Solution Approach 2:
The patent applies preliminary action by performing read extraction and contig assembly before final circRNA validation. By pre-processing the sequencing data into assembled contigs that represent potential circRNA structures, the method establishes a standardized basis for validation that reduces divergence across different bioinformatics approaches.
2Productivity
If multiple bioinformatics methods are applied to detect circRNAs, then more circRNA candidates are identified, but the complexity of validation increases
Solution Approach 1:
The patent extracts specific reads (spanning reads, split reads, and paired-end reads) that are relevant to circRNA detection from the full sequencing dataset. By taking out only the necessary reads for validation and assembling them into contigs, the method simplifies the validation process while maintaining high identification throughput across multiple bioinformatics methods.
3Measurement precision
If stringent validation criteria are applied to confirm circRNA junctions, then false positives are reduced, but more candidates are rejected for further analysis
Solution Approach 1:
The patent changes the validation parameters by using contig assembly and alignment to the reference genome as the validation criterion. Instead of relying on multiple different bioinformatics methods with varying parameters, the patent standardizes validation on contig-based evidence, allowing for consistent accuracy thresholds that do not excessively reduce candidate selection efficiency.
Data Source
AI summary
The present invention relates to bioinformatics methods of in silico validation of circRNA junctions and selection of circRNA junctions for further validation, more particularly detecting sequence reads supporting circRNA junction in a highly computationally efficient manner.