Substrate-Section-Specific Base Calling for Oligonucleotide Clusters
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing DNA sequencing systems face challenges in accurately distinguishing true light signals from wells of interest due to spatial crosstalk between adjacent wells, leading to sequencing errors and reduced base calling accuracy.
Innovation Solution
Implementing a system that uses deep convolutional neural networks to correct spatial crosstalk in cluster intensity data by applying sharpening masks and adaptive techniques to enhance base calling accuracy.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If deep convolutional neural networks with sharpening masks are applied to correct spatial crosstalk, then base calling accuracy is improved, but device complexity increases
Solution Approach 1:
The patent introduces an intermediary processing layer using deep convolutional neural networks with sharpening masks that mediate between the raw cluster intensity data and the final base calling decisions. This intermediary layer extracts and attenuates spatial crosstalk signals, separating them from true biological signals to improve measurement precision without requiring complete system redesign
Solution Approach 2:
The patent replaces traditional mechanical/optical signal separation methods with computational neural network-based processing. Instead of relying solely on physical optical isolation between wells, the system uses AI algorithms to computationally remove spatial crosstalk effects, achieving higher accuracy while managing complexity through software-based solutions
2Reliability
If spatial crosstalk attenuation methods are applied to reduce sequencing errors, then reliability is improved, but manufacturing precision requirements increase
Solution Approach 1:
The patent implements feedback mechanisms where the neural network continuously learns from sequencing data to refine its spatial crosstalk attenuation model. The system adapts to variations in well positioning and optical characteristics through training on reference data, allowing reliable sequencing even with manufacturing tolerances in well positioning without requiring ultra-precise manufacturing
Solution Approach 2:
The patent changes the approach from physically ensuring precise well positioning to computationally adjusting parameters to account for positioning variations. The neural network models and sharpening masks are trained to compensate for manufacturing precision variations, transforming the problem from a manufacturing constraint to a computational correction problem
Data Source
AI summary
The technology disclosed extracts intensities from sequencing images for base calling target clusters and attenuates spatial crosstalk from neighboring clusters. The technology disclosed accesses a particular section from a plurality of sections of an image output by a sensor, the particular section of the image including at least one pixel depicting intensity emission values from a target cluster and neighboring clusters located across the sensor, and convolves the particular section of the image with a corresponding convolution kernel in a plurality of convolution kernels, to generate a feature map comprising a plurality of feature values. The technology disclosed further assigns a corresponding feature value to the target cluster based on feature values in the plurality of feature values adjoining a center of the target cluster, and processes the corresponding feature value assigned to the target cluster, to base call the target cluster.