# US 12,119,088 B2 · Deep neural network-based sequencing

> Agents: this data is available as JSON from the patentgrep API and MCP server — see https://patentgrep.com/llms.txt

- Number: US 12,119,088 B2
- Document ID: US12119088B2
- Status: Granted
- Kind: B2
- Application number: 17/899,539
- Filed: 2022-08-30
- Published: 2023-01-05
- Granted: 2024-10-15
- Earliest claimed: 2019-03-21
- Assignees: Illumina, Inc.
- Applicants: Illumina, Inc.
- Inventors: Kishore Jaganathan; Anindita Dutta; Dorna Kashefhaghighi; John Randall Gobbel; Amirali Kia
- Web page: https://patentgrep.com/patent/US12119088B2

## CPC

- G16B 40/20: Supervised data analysis
- G06F 16/58: Retrieval characterised by using metadata, e.g. metadata not derived from the content or metadata generated manually
- G06F 16/907: Retrieval characterised by using metadata, e.g. metadata not derived from the content or metadata generated manually
- G06F 18/214: Generating training patterns; Bootstrap methods, e.g. bagging or boosting
- G06F 18/217: {Validation; Performance evaluation; Active pattern learning techniques}
- G06F 18/23: Clustering techniques
- G06F 18/23211: with adaptive number of clusters
- G06F 18/24: Classification techniques
- G06F 18/2415: based on parametric or probabilistic models, e.g. based on likelihood ratio or false acceptance rate versus a false rejection rate
- G06F 18/2431: Multiple classes
- G06N 3/04: Architecture, e.g. interconnection topology
- G06N 3/045: Combinations of networks
- G06N 3/0455: Auto-encoder networks; Encoder-decoder networks
- G06N 3/0464: Convolutional networks [CNN, ConvNet]
- G06N 3/08: Learning methods
- G06N 3/084: Backpropagation, e.g. using gradient descent
- G06N 3/09: Supervised learning
- G06N 7/01: Probabilistic graphical models, e.g. probabilistic networks
- G06V 10/267: {by performing operations on regions, e.g. growing, shrinking or watersheds}
- G06V 10/454: {Integrating the filters into a hierarchical structure, e.g. convolutional neural networks [CNN]}
- G06V 10/751: {Comparing pixel values or logical combinations thereof, or feature values having positional relevance, e.g. template matching}
- G06V 10/763: {Non-hierarchical techniques, e.g. based on statistics of modelling distributions}
- G06V 10/764: using classification, e.g. of video objects
- G06V 10/7715: {Feature extraction, e.g. by transforming the feature space, e.g. multi-dimensional scaling [MDS]; Mappings, e.g. subspace methods}
- G06V 10/7784: {based on feedback from supervisors}
- G06V 10/82: using neural networks
- G06V 10/993: {Evaluation of the quality of the acquired pattern}
- G06V 20/69: Microscopic objects, e.g. biological cells or cellular parts
- G16B 30/20: Sequence assembly
- G16B 40/00: ICT specially adapted for biostatistics; ICT specially adapted for bioinformatics-related machine learning or data mining, e.g. knowledge discovery or pattern finding
- G16B 40/10: Signal processing, e.g. from mass spectrometry [MS] or from PCR
- G06N 3/044: Recurrent networks, e.g. Hopfield networks
- G06N 3/048: Activation functions
- G06N 5/046: Forward inferencing; Production systems
- G06V 20/47: {Detecting features for summarising video content}
- G06V 2201/03: Recognition of patterns in medical or anatomical images
- G16B: BIOINFORMATICS, i.e. INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR GENETIC OR PROTEIN-RELATED DATA PROCESSING IN COMPUTATIONAL MOLECULAR BIOLOGY
- G06F: ELECTRIC DIGITAL DATA PROCESSING (computer systems based on specific computational models G06N)
- G06N: COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
- G06V: IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING

## Abstract

A system, a method and a non-transitory computer readable storage medium for base calling are described. The base calling method includes processing through a neural network first image data comprising images of clusters and their surrounding background captured by a sequencing system for one or more sequencing cycles of a sequencing run. The base calling method further includes producing a base call for one or more of the clusters of the one or more sequencing cycles of the sequencing run.

## Claims

### Claim 1

A sequencing system, comprising:
a receptacle coupled to a biosensor system, the biosensor system configured to comprise an array of light detectors, the biosensor system comprising a biosensor, and the biosensor comprising reaction sites configured to contain analytes;
an illumination system configured to direct excitation light toward the biosensor and illuminate the analytes in the reaction sites, wherein at least some of the analytes provide emission signals when illuminated; and
a system controller coupled to the receptacle and comprising an analysis module, the analysis module configured to:
obtain image data from the light detectors at each of a plurality of sequencing cycles of a sequencing run, wherein the image data is derived from the emission signals detected by the light detectors and comprises images of clusters and their surrounding background,
process the image data for each of the plurality of sequencing cycles through a neural network, and
produce a base call for one or more of the clusters at each of the plurality of sequencing cycles.

### Claim 2 (depends on 1)

The sequencing system of claim 1, wherein the analysis module is further configured to:
process a first input through a first neural network to produce a first output, wherein the first input comprises the image data; and
process a second input through a second neural network to produce a second output, wherein the second input comprises the image data and supplemental data, and wherein the second output identifies base calls for one or more of the clusters at each of the plurality of sequencing cycles.

### Claim 3 (depends on 2)

The sequencing system of claim 2, wherein the analysis module is further configured to generate the supplemental data by processing the first output to produce template data indicating one or more properties of respective portions of the image data.

### Claim 4 (depends on 1)

The sequencing system of claim 1, wherein the analysis module is further configured to:
process a first input through a first neural network to produce a first output, wherein the first input comprises the image data;
process the first output through a post-processor to produce template data indicating one or more properties of respective portions of the image data; and
process a second input through a second neural network to produce a second output, wherein the second input comprises the image data modified using the template data, and wherein the second output identifies base calls for one or more of the clusters at each of the plurality of sequencing cycles of the sequencing run.

### Claim 5 (depends on 4)

The sequencing system of claim 4, wherein the second input further comprises additional image data modified using the template data.

### Claim 6 (depends on 5)

The sequencing system of claim 5, wherein the additional image data comprises the images of the clusters and their surrounding background captured by the sequencing system for one or more additional sequencing cycles of the sequencing run.

### Claim 7 (depends on 5)

The sequencing system of claim 5, wherein modification using the template data comprises modifying intensities of one or more pixels in at least one of the image data or the additional image data based on an area weighting factor.

### Claim 8 (depends on 7)

The sequencing system of claim 7, further comprising calculating the area weighting factor for the one or more pixels in at least one of the image data or the additional image data based on how many subpixels in the template data that correspond to a pixel in the images of the image data or the additional image data contain parts of one or more of the clusters.

### Claim 9 (depends on 4)

The sequencing system of claim 4, wherein the template data comprises a template image, wherein the template image is at an upsampled, subpixel resolution.

### Claim 10 (depends on 9)

The sequencing system of claim 9, wherein each subpixel in the template image is identified as either background subpixel, cluster center subpixel, or cluster interior subpixel.

### Claim 11 (depends on 4)

The sequencing system of claim 4, wherein modification using the template data comprises assigning cluster intensities to subpixels in upsampled images that correspond to cluster center subpixels and cluster interior subpixels in a template image.

### Claim 12 (depends on 11)

The sequencing system of claim 11, further comprising:
upsampling the images of the clusters and their surrounding background to an upsampled, subpixel resolution to produce the upsampled images; and
assigning a background intensity to the subpixels in the upsampled images that correspond to background subpixels in the template image.

### Claim 13 (depends on 12)

The sequencing system of claim 12, wherein the background intensity has a zero value.

### Claim 14 (depends on 11)

The sequencing system of claim 11, wherein the cluster intensities are determined by interpolating intensities of pixels in an optical, pixel resolution.

### Claim 15 (depends on 4)

The sequencing system of claim 4, wherein modification using the template data comprises:
upsampling the images of the clusters and their surrounding background to an upsampled, subpixel resolution to produce upsampled images, and distributing an entire intensity of a pixel in an optical, pixel resolution among only those constituent subpixels of the pixel in the upsampled images that correspond to cluster center subpixels and cluster interior subpixels in a template image.

### Claim 16 (depends on 4)

The sequencing system of claim 4, wherein the template data identifies at least one of the one or more properties selected from a group consisting of: spatial distribution of the clusters, cluster shape, centers of the clusters and cluster boundary.

### Claim 17 (depends on 4)

The sequencing system of claim 4, further comprising calculating a quality of the base calls based on the second output.

### Claim 18 (depends on 1)

The sequencing system of claim 1, wherein the images of the clusters and their surrounding background are captured at an optical, pixel resolution.

### Claim 19 (depends on 1)

The sequencing system of claim 1, further comprising performing one or more sequencing cycles to capture the images of the clusters and their surrounding background.

### Claim 20 (depends on 1)

The sequencing system of claim 1, wherein each of the plurality of sequencing cycles generates a set of image data.