US 12,443,849 B2
Bus network for artificial intelligence-based base caller
Anindita Dutta, San Francisco, CA (US); Gery Vessere, Oakland, CA (US); Dorna Kashefhaghighi, Menlo Park, CA (US); Gavin Derek Parnaby, Laguna Niguel, CA (US); Kishore Jaganathan, San Francisco, CA (US); and Amirali Kia, San Mateo, CA (US)
Assigned to Illumina, Inc., San Diego, CA (US)
Filed by Illumina, Inc., San Diego, CA (US)
Filed on Feb. 19, 2021, as Appl. No. 17/180,513.
Claims priority of provisional application 62/979,411, filed on Feb. 20, 2020.
Claims priority of provisional application 62/979,399, filed on Feb. 20, 2020.
Prior Publication US 2021/0264267 A1, Aug. 26, 2021
Int. Cl. G06V 10/82 (2022.01); C12Q 1/6869 (2018.01); G06F 18/23 (2023.01); G06N 3/063 (2023.01); G06N 3/084 (2023.01); G06V 10/44 (2022.01); G06V 10/762 (2022.01); G06V 10/764 (2022.01); G06V 10/77 (2022.01); G16B 30/20 (2019.01)
CPC G06N 3/084 (2013.01) [G06F 18/23 (2023.01); G06N 3/063 (2013.01); G06V 10/454 (2022.01); G06V 10/762 (2022.01); G06V 10/764 (2022.01); G06V 10/7715 (2022.01); G06V 10/82 (2022.01); G16B 30/20 (2019.02); C12Q 1/6869 (2013.01)] 20 Claims
 
1. A system, comprising:
a sequencing instrument;
at least one processor; and
a non-transitory computer-readable medium comprising instructions that, when executed by the at least one processor, cause the system to:
capture, by one or more light detectors of the sequencing instrument at imaging events of sequencing cycles of a sequencing run, per-cycle sequencing image sets comprising signals from clusters of oligonucleotides;
process, through a spatial convolution network, a window of per-cycle sequencing image sets for a series of sequencing cycles of the sequencing cycles on a cycle-by-cycle basis by separately processing respective per-cycle sequencing image sets in the window of per-cycle sequencing image sets through respective spatial processing pipelines;
form, utilizing a bus network connected to the spatial convolutional network, skip buses between spatial convolution layers within respective sequences of spatial convolution layers of the respective spatial processing pipelines, the skip buses configured to cause respective per-cycle spatial feature map sets generated by two or more spatial convolution layers processing the respective per-cycle sequencing image sets in a particular sequence of spatial convolution layer for a particular sequencing cycle to combine into a combined per-cycle spatial feature map set, and provide the combined per-cycle spatial feature map set as input to another spatial convolution layer in the particular sequence of spatial convolution layer; and
generate, utilizing an output layer, base call predictions for the clusters of oligonucleotides depicted in the per-cycle sequencing image sets captured by the one or more light detectors of the sequencing instrument based on a final combined per-cycle spatial feature map set generated by the spatial convolution network.