MaxEnt and PWM Evaluator for Pre-mRNA Splicing Prediction
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing methods for predicting branchpoint mutations in RNA splicing are inaccurate due to reliance on wet lab techniques and computational models based on hypothetical assumptions, which are time-consuming and labor-intensive, and do not account for the complex mechanisms of branchpoint and splice site selection, limiting their application in clinical diagnostics.
Innovation Solution
A system and method using a MaxEnt tool and Position Weight Matrix (PWM) evaluator to predict the effect of genomic variations on pre-mRNA splicing, capable of operating in resource-constrained environments, accurately identifying pathogenic variants and their pathogenicity by analyzing candidate variants in splice acceptor and branch site regions.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If wet lab techniques are used for branchpoint prediction, then prediction accuracy is improved, but time consumption and labor intensity increase significantly
Solution Approach 1:
The patent creates computational models that replicate the function of wet lab techniques through in-silico simulations. The MaxEnt tool and PWM evaluator copy the predictive capability of experimental methods while eliminating the need for physical laboratory work, achieving both accuracy and efficiency.
Solution Approach 2:
The patent replaces mechanical/wet lab systems with computational algorithms. The MaxEnt (Maximum Entropy) tool and Position Weight Matrix evaluator substitute physical experimental procedures with mathematical models, transforming a labor-intensive process into an automated computational pipeline.
2Productivity
If computational models based on hypothetical assumptions are used, then speed is improved, but prediction accuracy deteriorates
Solution Approach 1:
The patent changes the parameters and assumptions underlying computational models. Instead of using hypothetical assumptions, the MaxEnt tool and PWM evaluator are trained on experimentally validated data, adjusting the statistical parameters to reflect real biological patterns while maintaining computational speed.
Solution Approach 2:
The patent incorporates feedback mechanisms where computational predictions are continuously refined based on experimental validation results. The models learn from discrepancies between predicted and observed branchpoints, iteratively improving accuracy while maintaining efficient computation.
3Ease of operation
If existing computational tools are used for splicing defect prediction, then ease of operation is improved, but reliability deteriorates due to lack of branchpoint selection mechanism modeling
Solution Approach 1:
The patent segments the splicing prediction problem into distinct functional components: the MaxEnt tool for initial branchpoint identification, the PWM evaluator for splice site assessment, and the integrated pipeline for comprehensive prediction. This modular segmentation maintains ease of operation while improving reliability through specialized processing at each stage.
Solution Approach 2:
The patent creates a universal prediction system that handles multiple aspects of splicing prediction (branchpoint selection, splice site recognition, variant pathogenicity assessment) within a single integrated pipeline. The MaxEnt tool and PWM evaluator work together to provide multi-functional analysis, improving reliability without complicating user operation.
Data Source
Figure 1
Figure 2
Figure 3A
AI summary
This disclosure relates generally to method and system for predicting effect of genomic variations on pre-mRNA splicing. The method include receiving genomic position information of at least one candidate variant, gene transcripts and genomic coordinates information of the gene transcripts; classifying the at least one candidate variant into one of a splice acceptor site region and a branch site region based on the coordinates information of the gene transcripts and the genomic position information of at least one candidate variant; evaluating effect of the at least one candidate variant on pre-mRNA splicing, based on a classified region from the classification of the at least one candidate variant and predicting pathogenicity of the at least one candidate variant based on the evaluated effect of the at least one candidate variant on the pre-mRNA splicing.