John Randall Gobbel

US Patent:

20200327377, Oct 15, 2020

Filed:

Mar 20, 2020

Appl. No.:

16/826134

Inventors:

- San Diego CA, US
John Randall GOBBEL - Brisbane CA, US
Amirali KIA - San Mateo CA, US

Assignee:

Illumina, Inc. - San Diego CA

International Classification:

G06K 9/62
G06N 3/08
G06N 7/00

Abstract:

The technology disclosed assigns quality scores to bases called by a neural network-based base caller by (i) quantizing classification scores of predicted base calls produced by the neural network-based base caller in response to processing training data during training, (ii) selecting a set of quantized classification scores, (iii) for each quantized classification score in the set, determining a base calling error rate by comparing its predicted base calls to corresponding ground truth base calls, (iv) determining a fit between the quantized classification scores and their base calling error rates, and (v) correlating the quality scores to the quantized classification scores based on the fit.

US Patent:

20200302224, Sep 24, 2020

Filed:

Mar 21, 2020

Appl. No.:

16/826168

Inventors:

- San Diego CA, US
Anindita DUTTA - Brisbane CA, US
Dorna KASHEFHAGHIGHI - Menlo Park CA, US
John Randall GOBBEL - Brisbane CA, US
Amirali KIA - San Mateo CA, US

Assignee:

Illumina, Inc. - San Diego CA

International Classification:

G06K 9/62
G06N 3/08
G06F 16/907

Abstract:

The technology disclosed processes a first input through a first neural network and produces a first output. The first input comprises first image data derived from images of analytes and their surrounding background captured by a sequencing system for a sequencing run. The technology disclosed processes the first output through a post-processor and produces metadata about the analytes and their surrounding background. The technology disclosed processes a second input through a second neural network and produces a second output. The second input comprises third image data derived by modifying second image data based on the metadata. The second image data is derived from the images of the analytes and their surrounding background. The second output identifies base calls for one or more of the analytes at one or more sequencing cycles of the sequencing run.

US Patent:

20200302297, Sep 24, 2020

Filed:

Mar 20, 2020

Appl. No.:

16/826126

Inventors:

- San Diego CA, US
John Randall GOBBEL - Brisbane CA, US
Amirali KIA - San Mateo CA, US

Assignee:

Illumina, Inc. - San Diego CA

International Classification:

G06N 3/08
G06N 3/04

Abstract:

The technology disclosed processes input data through a neural network and produces an alternative representation of the input data. The input data includes per-cycle image data for each of one or more sequencing cycles of a sequencing run. The per-cycle image data depicts intensity emissions of one or more analytes and their surrounding background captured at a respective sequencing cycle. The technology disclosed processes the alternative representation through an output layer and producing an output and base calls one or more of the analytes at one or more of the sequencing cycles based on the output.