Patents by Inventor Kai-How FARH
Kai-How FARH has filed for patents to protect the following inventions. This listing includes patent applications that are pending as well as patents that have already been granted by the United States Patent and Trademark Office (USPTO).
-
Patent number: 11397889Abstract: The technology disclosed relates to constructing a convolutional neural network-based classifier for variant classification. In particular, it relates to training a convolutional neural network-based classifier on training data using a backpropagation-based gradient update technique that progressively match outputs of the convolutional network network-based classifier with corresponding ground truth labels. The convolutional neural network-based classifier comprises groups of residual blocks, each group of residual blocks is parameterized by a number of convolution filters in the residual blocks, a convolution window size of the residual blocks, and an atrous convolution rate of the residual blocks, the size of convolution window varies between groups of residual blocks, the atrous convolution rate varies between groups of residual blocks. The training data includes benign training examples and pathogenic training examples of translated sequence pairs generated from benign variants and pathogenic variants.Type: GrantFiled: October 15, 2018Date of Patent: July 26, 2022Assignee: Illumina, Inc.Inventors: Kishore Jaganathan, Kai-How Farh, Sofia Kyriazopoulou Panagiotopoulou, Jeremy Francis McRae
-
Patent number: 11386324Abstract: The technology disclosed relates to constructing a convolutional neural network-based classifier for variant classification. In particular, it relates to training a convolutional neural network-based classifier on training data using a backpropagation-based gradient update technique that progressively match outputs of the convolutional neural network-based classifier with corresponding ground truth labels. The convolutional neural network-based classifier comprises groups of residual blocks, each group of residual blocks is parameterized by a number of convolution filters in the residual blocks, a convolution window size of the residual blocks, and an atrous convolution rate of the residual blocks, the size of convolution window varies between groups of residual blocks, the atrous convolution rate varies between groups of residual blocks. The training data includes benign training examples and pathogenic training examples of translated sequence pairs generated from benign variants and pathogenic variants.Type: GrantFiled: January 27, 2020Date of Patent: July 12, 2022Assignee: Illumina, Inc.Inventors: Hong Gao, Kai-How Farh, Laksshman Sundaram, Jeremy Francis McRae
-
Publication number: 20220164710Abstract: Systems, computer-implemented methods, and non-transitory computer readable media are provided for sharing medical data. The disclosed systems may be configured to create a first workgroup having a first knowledgebase. This first knowledgebase may be federated with a common knowledgebase, and with a second knowledgebase of a second workgroup. At least one of the first knowledgebase, common knowledgebase, and second knowledgebase may be configured to store data items comprising associations, signs, and evidence. The signs may comprise measurements and contexts, and the associations may describe the relationships between the measurements and contexts. The evidence may support these associations. The disclosed systems may be configured to receive a request from a user in the first workgroup, retrieve matching data items, and optionally then output to the user at least some of the retrieved matching data items. The request may comprise at least one of a first association and a first measurement.Type: ApplicationFiled: February 7, 2022Publication date: May 26, 2022Inventors: Kai-How Farh, Donavan Cheng, John Casey Shon, Jorg Hakenberg, Eugene Bolotin, James Geaney, Hong Gao, Pam Cheng, Inderjit Singh, Daniel Roche, Milan Karangutkar
-
Patent number: 11315016Abstract: The technology disclosed relates to constructing a convolutional neural network-based classifier for variant classification. In particular, it relates to training a convolutional neural network-based classifier on training data using a backpropagation-based gradient update technique that progressively match outputs of the convolutional neural network-based classifier with corresponding ground truth labels. The convolutional neural network-based classifier comprises groups of residual blocks, each group of residual blocks is parameterized by a number of convolution filters in the residual blocks, a convolution window size of the residual blocks, and an atrous convolution rate of the residual blocks, the size of convolution window varies between groups of residual blocks, the atrous convolution rate varies between groups of residual blocks. The training data includes benign training examples and pathogenic training examples of translated sequence pairs generated from benign variants and pathogenic variants.Type: GrantFiled: October 15, 2018Date of Patent: April 26, 2022Assignee: Illumina, Inc.Inventors: Laksshman Sundaram, Kai-How Farh, Hong Gao, Samskruthi Reddy Padigepati, Jeremy Francis McRae
-
Patent number: 11244246Abstract: Systems, computer-implemented methods, and non-transitory computer readable media are provided for sharing medical data. The disclosed systems may be configured to create a first workgroup having a first knowledgebase. This first knowledgebase may be federated with a common knowledgebase, and with a second knowledgebase of a second workgroup. At least one of the first knowledgebase, common knowledgebase, and second knowledgebase may be configured to store data items comprising associations, signs, and evidence. The signs may comprise measurements and contexts, and the associations may describe the relationships between the measurements and contexts. The evidence may support these associations. The disclosed systems may be configured to receive a request from a user in the first workgroup, retrieve matching data items, and optionally then output to the user at least some of the retrieved matching data items. The request may comprise at least one of a first association and a first measurement.Type: GrantFiled: February 24, 2020Date of Patent: February 8, 2022Assignee: ILLUMINA, INC.Inventors: Kai-How Farh, Donavan Cheng, John Shon, Jorg Hakenberg, Eugene Bolotin, James Casey Geaney, Hong Gao, Pam Cheng, Inderjit Singh, Daniel Roche, Milan Karangutkar
-
Publication number: 20220027388Abstract: Derivation and use of pathogenicity scores for gene variants are described herein. Applications, uses, and variations of the pathogenicity scoring process include, but are not limited to, the derivation and use of thresholds to characterize a variant as pathogenic or benign, the estimation of selection effects associated with a gene variant, the estimation of genetic disease prevalence using pathogenicity scores, and the recalibration of methods used to assess pathogenicity scores.Type: ApplicationFiled: July 21, 2021Publication date: January 27, 2022Inventors: Hong Gao, Kai-How Farh, Jeremy Francis McRae
-
Publication number: 20220028485Abstract: Derivation and use of pathogenicity scores for gene variants are described herein. Applications, uses, and variations of the pathogenicity scoring process include, but are not limited to, the derivation and use of thresholds to characterize a variant as pathogenic or benign, the estimation of selection effects associated with a gene variant, the estimation of genetic disease prevalence using pathogenicity scores, and the recalibration of methods used to assess pathogenicity scores.Type: ApplicationFiled: July 21, 2021Publication date: January 27, 2022Inventors: Hong Gao, Kai-How Farh, Jeremy Francis McRae
-
Publication number: 20200380412Abstract: Systems, computer-implemented methods, and non-transitory computer readable media are provided for sharing medical data. The disclosed systems may be configured to create a first workgroup having a first knowledgebase. This first knowledgebase may be federated with a common knowledgebase, and with a second knowledgebase of a second workgroup. At least one of the first knowledgebase, common knowledgebase, and second knowledgebase may be configured to store data items comprising associations, signs, and evidence. The signs may comprise measurements and contexts, and the associations may describe the relationships between the measurements and contexts. The evidence may support these associations. The disclosed systems may be configured to receive a request from a user in the first workgroup, retrieve matching data items, and optionally then output to the user at least some of the retrieved matching data items. The request may comprise at least one of a first association and a first measurement.Type: ApplicationFiled: February 24, 2020Publication date: December 3, 2020Inventors: Kai-How FARH, Donavan CHENG, John SHON, Jorg HAKENBERG, Eugene BOLOTIN, James Casey GEANEY, Hong GAO, Pam CHENG, Inderjit SINGH, Daniel ROCHE, Milan KARANGUTKAR
-
Publication number: 20200279157Abstract: The technology disclosed relates to constructing a convolutional neural network-based classifier for variant classification. In particular, it relates to training a convolutional neural network-based classifier on training data using a backpropagation-based gradient update technique that progressively match outputs of the convolutional neural network-based classifier with corresponding ground truth labels. The convolutional neural network-based classifier comprises groups of residual blocks, each group of residual blocks is parameterized by a number of convolution filters in the residual blocks, a convolution window size of the residual blocks, and an atrous convolution rate of the residual blocks, the size of convolution window varies between groups of residual blocks, the atrous convolution rate varies between groups of residual blocks. The training data includes benign training examples and pathogenic training examples of translated sequence pairs generated from benign variants and pathogenic variants.Type: ApplicationFiled: January 27, 2020Publication date: September 3, 2020Applicant: Illumina, Inc.Inventors: Hong GAO, Kai-How FARH, Laksshman SUNDARAM, Jeremy Francis McRAE
-
Publication number: 20200251183Abstract: The technology disclosed presents a deep learning-based framework, which identifies sequence patterns that cause sequence-specific errors (SSEs). Systems and methods train a variant filter on large-scale variant data to learn causal dependencies between sequence patterns and false variant calls. The variant filter has a hierarchical structure built on deep neural networks such as convolutional neural networks and fully-connected neural networks. Systems and methods implement a simulation that uses the variant filter to test known sequence patterns for their effect on variant filtering. The premise of the simulation is as follows: when a pair of a repeat pattern under test and a called variant is fed to the variant filter as part of a simulated input sequence and the variant filter classifies the called variant as a false variant call, then the repeat pattern is considered to have caused the false variant call and identified as SSE-causing.Type: ApplicationFiled: July 8, 2019Publication date: August 6, 2020Applicant: Illumina, Inc.Inventors: Dorna KASHEFHAGHIGHI, Amirali KIA, Kai-How FARH
-
Patent number: 10607156Abstract: Systems, computer-implemented methods, and non-transitory computer readable media are provided for sharing medical data. The disclosed systems may be configured to create a first workgroup having a first knowledgebase. This first knowledgebase may be federated with a common knowledgebase, and with a second knowledgebase of a second workgroup. At least one of the first knowledgebase, common knowledgebase, and second knowledgebase may be configured to store data items comprising associations, signs, and evidence. The signs may comprise measurements and contexts, and the associations may describe the relationships between the measurements and contexts. The evidence may support these associations. The disclosed systems may be configured to receive a request from a user in the first workgroup, retrieve matching data items, and optionally then output to the user at least some of the retrieved matching data items. The request may comprise at least one of a first association and a first measurement.Type: GrantFiled: August 22, 2017Date of Patent: March 31, 2020Assignee: ILLUMINA, INC.Inventors: Kai-How Farh, Donavan Cheng, John Shon, Jorg Hakenberg, Eugene Bolotin, James Casey Geaney, Hong Gao, Pam Cheng, Inderjit Singh, Daniel Roche, Milan Karangutkar
-
Publication number: 20200065675Abstract: The technology disclosed relates to constructing a convolutional neural network-based classifier for variant classification. In particular, it relates to training a convolutional neural network-based classifier on training data using a backpropagation-based gradient update technique that progressively match outputs of the convolutional network network-based classifier with corresponding ground truth labels. The convolutional neural network-based classifier comprises groups of residual blocks, each group of residual blocks is parameterized by a number of convolution filters in the residual blocks, a convolution window size of the residual blocks, and an atrous convolution rate of the residual blocks, the size of convolution window varies between groups of residual blocks, the atrous convolution rate varies between groups of residual blocks. The training data includes benign training examples and pathogenic training examples of translated sequence pairs generated from benign variants and pathogenic variants.Type: ApplicationFiled: October 15, 2018Publication date: February 27, 2020Applicant: Illumina, Inc.Inventors: Laksshman Sundaram, Kai-How Farh, Hong Gao, Samskruthi Reddy Padigepati, Jeremy Francis McRae
-
Patent number: 10558915Abstract: The technology disclosed relates to constructing a convolutional neural network-based classifier for variant classification. In particular, it relates to training a convolutional neural network-based classifier on training data using a backpropagation-based gradient update technique that progressively match outputs of the convolutional neutral network-based classifier with corresponding ground truth labels. The convolutional neural network-based classifier comprises groups of residual blocks, each group of residual blocks is parameterized by a number of convolution filters in the residual blocks, a convolution window size of the residual blocks, and an atrous convolution rate of the residual blocks, the size of convolution window varies between groups of residual blocks, the atrous convolution rate varies between groups of residual blocks. The training data includes benign training examples and pathogenic training examples of translated sequence pairs generated from benign variants and pathogenic variants.Type: GrantFiled: May 15, 2019Date of Patent: February 11, 2020Assignee: Illumina, Inc.Inventors: Hong Gao, Kai-How Farh, Laksshman Sundaram, Jeremy Francis McRae
-
Patent number: 10540591Abstract: The technology disclosed includes systems and methods to reduce overfitting of neural network-implemented models that process sequences of amino acids and accompanying position frequency matrices. The system generates supplemental training example sequence pairs, labelled benign, that include a start location, through a target amino acid location, to an end location. A supplemental sequence pair supplements a pathogenic or benign missense training example sequence pair. It has identical amino acids in a reference and an alternate sequence of amino acids. The system includes logic to input with each supplemental sequence pair a supplemental training position frequency matrix (PFM) that is identical to the PFM of the benign or pathogenic missense at the matching start and end location. The system includes logic to attenuate the training influence of the training PFMs during training the neural network-implemented model by including supplemental training example PFMs in the training data.Type: GrantFiled: May 8, 2019Date of Patent: January 21, 2020Assignee: Illumina, Inc.Inventors: Hong Gao, Kai-How Farh, Samskruthi Reddy Padigepati
-
Publication number: 20200019859Abstract: We disclose computational models that alleviate the effects of human ascertainment biases in curated pathogenic non-coding variant databases by generating pathogenicity scores for variants occurring in the promoter regions (referred to herein as promoter single nucleotide variants (pSNVs)). We train deep learning networks (referred to herein as pathogenicity classifiers) using a semi-supervised approach to discriminate between a set of labeled benign variants and an unlabeled set of variants that were matched to remove biases.Type: ApplicationFiled: September 20, 2019Publication date: January 16, 2020Applicant: Illumina, Inc.Inventors: Sofia KYRIAZOPOULOU PANAGIOTOPOULOU, Kai-How FARH
-
Patent number: 10509795Abstract: Systems, computer-implemented methods, and non-transitory computer readable media are provided for determining related ontological data. The disclosed systems may be configured to receive a first ontology and a second ontology, the first ontology and the second ontology comprising hierarchically organized ontological data. The disclosed systems may also be configured to receive an indication that a first ontological datum in the first ontology is equivalent to a second ontological datum in the second ontology, and a query for ontological data related to a third ontological datum subordinate to the first ontological datum. The disclosed systems may be configured to determine a first semantic distance between the third ontological datum and a fourth ontological datum in the second ontology satisfies a semantic distance criterion, and output the fourth ontological datum based on the determination that the first semantic distance satisfies the semantic distance criterion.Type: GrantFiled: August 22, 2017Date of Patent: December 17, 2019Assignee: ILLUMINA, INC.Inventors: Kai-How Farh, Jorg Hakenberg, Milan Karangutkar, Wenwu Cui, Hong Gao
-
Patent number: 10423861Abstract: The technology disclosed relates to constructing a convolutional neural network-based classifier for variant classification. In particular, it relates to training a convolutional neural network-based classifier on training data using a backpropagation-based gradient update technique that progressively match outputs of the convolutional network network-based classifier with corresponding ground truth labels. The convolutional neural network-based classifier comprises groups of residual blocks, each group of residual blocks is parameterized by a number of convolution filters in the residual blocks, a convolution window size of the residual blocks, and an atrous convolution rate of the residual blocks, the size of convolution window varies between groups of residual blocks, the atrous convolution rate varies between groups of residual blocks. The training data includes benign training examples and pathogenic training examples of translated sequence pairs generated from benign variants and pathogenic variants.Type: GrantFiled: October 15, 2018Date of Patent: September 24, 2019Assignee: Illumina, Inc.Inventors: Hong Gao, Kai-How Farh, Laksshman Sundaram, Jeremy Francis McRae
-
Publication number: 20190266493Abstract: The technology disclosed includes systems and methods to reduce overfitting of neural network-implemented models that process sequences of amino acids and accompanying position frequency matrices. The system generates supplemental training example sequence pairs, labelled benign, that include a start location, through a target amino acid location, to an end location. A supplemental sequence pair supplements a pathogenic or benign missense training example sequence pair. It has identical amino acids in a reference and an alternate sequence of amino acids. The system includes logic to input with each supplemental sequence pair a supplemental training position frequency matrix (PFM) that is identical to the PFM of the benign or pathogenic missense at the matching start and end location. The system includes logic to attenuate the training influence of the training PFMs during training the neural network-implemented model by including supplemental training example PFMs in the training data.Type: ApplicationFiled: May 8, 2019Publication date: August 29, 2019Applicant: Illumina, Inc.Inventors: Hong GAO, Kai-How FARH, Samskruthi REDDY PADIGEPATI
-
Patent number: D869489Type: GrantFiled: May 25, 2018Date of Patent: December 10, 2019Assignee: Illumina, Inc.Inventors: Kai-How Farh, Donavan Cheng, Andrew Warren, Ian D. Patrick
-
Patent number: D875773Type: GrantFiled: September 11, 2019Date of Patent: February 18, 2020Assignee: Illumina, Inc.Inventors: Kai-How Farh, Donavan Cheng, Andrew Warren, Ian D. Patrick