Nucleic Acid Base Calling from Weak-Signal Image Clustering
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing sequencing technologies face issues with incomplete detection of image features, leading to loss of throughput in base calling due to incompleteness of constructed templates, particularly with weak reaction signals and low sensitivity in signal detection.
Innovation Solution
A method and apparatus for determining a base sequence by processing images at basic units, detecting base types at each unit, and clustering similar sequences to improve accuracy and throughput, using a processing module, detection module, and determination module to enhance signal detection.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If template construction based on image feature identification is used for base calling, then base calling can be performed, but throughput is lost due to incomplete identification of image features and incompleteness of constructed templates
Solution Approach 1:
The patent segments the image processing into multiple passes: first identifying strong signal features to construct an initial template, then using this template to guide identification of weaker features in subsequent passes. This segmentation allows the system to handle incomplete initial identification without losing throughput, as the template progressively improves to capture more features including weak signals.
Solution Approach 2:
The patent performs preliminary template construction using only strong signal features before attempting to detect weak features. This preliminary action creates a reliable foundation template that guides subsequent detection, ensuring that weak features are identified in the context of a already-established template structure, thereby preventing throughput loss from incomplete identification.
2Device complexity
If traditional base calling methods are used, then processing is simpler, but detection sensitivity is low and weak reaction signals are missed
Solution Approach 1:
The patent introduces a template as an intermediary structure that mediates between the raw image data and the base calling process. This template acts as a reference framework that enhances the detection of weak signals by providing expected feature patterns, thereby improving detection sensitivity without requiring fundamentally complex processing changes to the base calling mechanism itself.
Solution Approach 2:
The patent replaces direct mechanical/direct detection of all features simultaneously with a two-stage process: first constructing a template from strong features, then using this template to guide detection of weak features. This substitution of the direct detection mechanism with a template-guided approach improves sensitivity to weak signals while keeping processing complexity manageable.
3Loss of information
If all image features are attempted to be identified simultaneously, then complete template construction is achieved, but processing complexity increases and weak signals are lost in noise
Solution Approach 1:
The patent segments the feature identification process into distinct stages: first identifying strong signal features to build an initial template, then using this template to guide identification of weak features. This segmentation prevents the complexity of simultaneous identification while ensuring complete feature detection, as each stage builds upon the previous one with manageable complexity at each step.
Solution Approach 2:
The patent performs preliminary identification of strong features and template construction before attempting to detect weak features. This preliminary action simplifies the overall processing by establishing a reliable template framework first, which then serves as a guide for detecting weaker signals, thereby achieving complete feature detection without the overwhelming complexity of simultaneous identification.
Applied Scientific Principles
This section explains which scientific principles are used to turn an abstract innovation direction into a practical engineering solution.
Function Achieved in This Case
Enhances sequencing accuracy and throughput by reducing the loss of target signal detection from weak reaction signals and low sensitivity, improving the overall sequencing process.
Implementation Method 1
detecting a reaction signal from a surface by using an optical imaging system to acquire an image, where the reaction signal corresponding to a specific chemical feature may be presented as a spot or a point in the image
Data Source
AI summary
The present application discloses a method and apparatus for determining a base sequence of a nucleic acid template, a device, and a medium, and generally relates to the field of data processing. The method includes: processing an image including a feature corresponding to the nucleic acid template, including: determining a signal intensity at each basic unit position in the image, where the image includes a plurality of basic units, the size of the feature corresponding to the nucleic acid template in the image is represented as one or more basic units, and the size of one basic unit is less than or equal to the size of one pixel of the image; detecting, based on the signal intensity at each basic unit position, the type of one or more bases incorporated into the nucleic acid template corresponding to the basic unit position to determine a detected base sequence at each basic unit position; and clustering, based on a similarity between the detected base sequence at each basic unit position and detected base sequences at surrounding basic unit positions thereof, the detected base sequences or the basic unit positions to determine a portion of the base sequence of the nucleic acid template. The present application can improve the sequencing accuracy and sequencing throughput.