Equalizer-Based Base Calling for Sequencing Image Crosstalk
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 an equalizer-based approach that generates lookup tables (LUTs) to correct spatial crosstalk by training an equalizer using least square estimation, maximizing the signal-to-noise ratio through coefficient learning and applying these coefficients to pixel intensities to attenuate unwanted light signals.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If spatial crosstalk correction is not applied, then the system complexity remains low, but base calling accuracy deteriorates due to inability to distinguish true light signals from adjacent wells
Solution Approach 1:
The equalizer coefficients are trained in advance using least square estimation to maximize signal-to-noise ratio. This preliminary training phase creates lookup tables that are then applied during actual base calling, allowing the system to correct spatial crosstalk without adding real-time computational complexity
Solution Approach 2:
An equalizer component is introduced as an intermediary between the raw sensor signals and the base calling process. This equalizer uses pre-trained coefficients to filter and correct spatial crosstalk from adjacent wells, enabling accurate signal distinction without requiring complex real-time processing
2Productivity
If equalizer training is performed offline, then base calling speed improves, but the adaptability to different sequencing conditions deteriorates
Solution Approach 1:
The system implements dynamic adaptability by allowing equalizer retraining when sequencing conditions change. The equalizer can be retrained with new training data corresponding to different sequencing runs or conditions, enabling the system to adapt to varying scenarios while maintaining fast base calling through the use of pre-trained coefficients
3Reliability
If more pixels are used to capture light signals, then signal detection capability improves, but spatial crosstalk from adjacent wells worsens
Solution Approach 1:
The system converts the harmful spatial crosstalk from adjacent wells into useful information for correction. By using signals from multiple pixels including those affected by crosstalk, the equalizer can learn the crosstalk pattern and subtract it, effectively converting the harmful interference into correctable data that improves overall signal detection accuracy
Data Source
AI summary
The technology disclosed relates to equalizer-based intensity correction for base calling. In particular, the technology disclosed relates to accessing an image whose pixels depict intensity emissions from a target cluster and intensity emissions from additional adjacent clusters, selecting a lookup table that contains pixel coefficients that are configured to increase a signal-to-noise ratio, applying the pixel coefficients to intensity values of the pixels in the image to produce an output, and base calling the target cluster based on the output.