Healthcare Economics Consultant at UnitedHealthcare
Location:
Gaithersburg, Maryland
Industry:
Insurance
Work:
UnitedHealthcare since Jun 2011
Healthcare Economics Consultant
Coventry Health Care Dec 2010 - Jun 2011
Actuarial Intern
Industrial and Commercial Bank of China Jul 2000 - Nov 2005
Business Analyst
Education:
University of Iowa 2008 - 2010
M.S., Statistics w/ Emphasis on Actuarial Science
Nankai University 1996 - 2000
B.S., Accounting
EMC Corporation Irvine, CA Jun 2011 to Sep 2011 Server QA InternAlipay.com Hangzhou, China Mar 2010 to May 2010 Software Configuration Management Engineer (Intern)
Education:
University of California, Irvine Irvine, CA 2010 to 2012 MS in Information and Computer ScienceCommunication University of China 2006 to 2010 BE in Computer Science and Technology
Skills:
Languages: Java, Python, C#; Internet: ASP.NET, IIS, Java Servlet, JSP, HTML; Databases: SQL, MS SQL Server, DB2, MySQL; Operating Systems: Windows, Linux, Mac; Experience in Android development; Experience in Arduino development; Knowledge of software configuration management; Understanding of HCI
A composition of material for making a glow-in-the-dark powder and glaze as well as a process for manufacturing a glow-in-the-dark ceramic. The glow-in-the-dark powder consists of aluminum oxide, strontium carbonate, rare earth, and carbon. The glow-in-the-dark glaze consists of red lead, quartz, zinc oxide, glaze melting block, calcite, Suzhou earth, stannic oxide, limestone, and the glow-in-the-dark powder. The glow-in-the dark powder is first sintered and then grinded to fine powder. The powder is then added to the glow-in-the-dark glaze and stirred vigorously to form a paste that can be applied to ceramic products which are to be sintered to form the final glow-in-the-dark ceramic products.
Multi-Channel Protein Voxelization To Predict Variant Pathogenicity Using Deep Convolutional Neural Networks
- San Diego CA, US - Cambridge, GB Hong GAO - Palo Alto CA, US
Assignee:
Illumina, Inc. - San Diego CA Illumina Cambridge Limited - Cambridge
International Classification:
G16B 40/20 G16B 15/20 G16B 30/00
Abstract:
A system includes at least a voxelizer, an alternative allele encoder, an evolutionary conservation encoder, and a convolutional neural network. The voxelizer accesses a three-dimensional structure of a reference amino acid sequence of a protein and fits a three-dimensional grid of voxels on atoms in the three-dimensional structure on an amino acid-basis to generate amino acid-wise distance channels. The alternative allele encoder encodes an alternative allele sequence to each voxel in the three-dimensional grid of voxels. The evolutionary conservation encoder encodes an evolutionary conservation sequence to each voxel in the three-dimensional grid of voxels. The convolutional neural network applies three-dimensional convolutions to a tensor that includes the amino acid-wise distance channels encoded with the alternative allele sequence and respective evolutionary conservation sequences and determines a pathogenicity of a variant nucleotide based at least in part on the tensor.
The technology disclosed relates to efficiently determining which atoms in a protein are nearest to voxels in a grid. The atoms have three-dimensional (3D) atom coordinates, and the voxels have 3D voxel coordinates. The technology disclosed generates an atom-to-voxels mapping that maps, to each of the atoms, a containing voxel selected based on matching 3D atom coordinates of a particular atom of the protein to the 3D voxel coordinates in the grid. The technology disclosed generates a voxel-to-atoms mapping that maps, to each of the voxels, a subset of the atoms. The subset of the atoms mapped to a particular voxel in the grid includes those atoms in the protein that are mapped to the particular voxel by the atom-to-voxels mapping. The technology disclosed includes using the voxel-to-atoms mapping to determine, for each of the voxels, a nearest atom in the protein.
Variant Pathogenicity Prediction Using Neural Network
The technology disclosed relates to constructing a computer-implemented method for variant classification. In particular, the method includes using a pathogenicity prediction neural network to process as input, (i) a reference protein sequence that has a first chain of amino acids with at least twenty amino acids, (ii) an alternative protein sequence aligned with the reference sequence, where the alternative protein sequence has a second chain of amino acids with at least twenty amino acids, and the first and second chains of amino acids differ by a variant amino acid caused by a nucleotide substitution, and (iii) a primate conservation profile generated using a primate cross-species multiple sequence alignment that aligns the reference protein sequence with other protein sequences from primate species. The method further includes based on the processing of the input by the neural network, generating as output a pathogenicity prediction for the nucleotide substitution.
Transfer Learning-Based Use Of Protein Contact Maps For Variant Pathogenicity Prediction
The technology disclosed relates to a variant pathogenicity prediction network. The variant pathogenicity classifier includes memory, a variant encoding sub-network, a protein contact map generation sub-network, and a pathogenicity scoring sub-network. The memory stores a reference amino acid sequence of a protein, and an alternative amino acid sequence of the protein that contains a variant amino acid caused by a variant nucleotide. The variant encoding sub-network is configured to process the alternative amino acid sequence, and generate a processed representation of the alternative amino acid sequence. The protein contact map generation sub-network is configured to process the reference amino acid sequence and the processed representation of the alternative amino acid sequence, and generate a protein contact map of the protein. The pathogenicity scoring sub-network is configured to process the protein contact map, and generate a pathogenicity indication of the variant amino acid.
Deep Learning-Based Use Of Protein Contact Maps For Variant Pathogenicity Prediction
The technology disclosed relates to a variant pathogenicity classifier. The variant pathogenicity classifier comprises memory and runtime logic. The memory stores (i) a reference amino acid sequence of a protein, (ii) an alternative amino acid sequence of the protein that contains a variant amino acid caused by a variant nucleotide, and (iii) a protein contact map of the protein. The runtime logic has access to the memory, and is configured to provide (i) the reference amino acid sequence, (ii) the alternative amino acid sequence, and (iii) the protein contact map as input to a first neural network, and to cause the first neural network to generate a pathogenicity indication of the variant amino acid as output in response to processing (i) the reference amino acid sequence, (ii) the alternative amino acid sequence, and (iii) the protein contact map.
Federated Systems And Methods For Medical Data Sharing
- San Diego CA, US Donavan CHENG - Foster City CA, US John SHON - Portola Valley CA, US Jorg HAKENBERG - Foster City CA, US Eugene BOLOTIN - San Diego CA, US James Casey GEANEY - Oakland CA, US Hong GAO - San Diego CA, US Pam CHENG - San Diego CA, US Inderjit SINGH - San Diego CA, US Daniel ROCHE - San Diego CA, US Milan KARANGUTKAR - Santa Clara CA, US
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.
Deep Convolutional Neural Networks For Variant Classification
- San Diego CA, US Hong Gao - Palo Alto CA, US Samskruthi Reddy Padigepati - Sunnyvale CA, US Jeremy Francis McRae - Hayward CA, US
Assignee:
Illumina, Inc. - San Diego CA
International Classification:
G06N 3/08 G06N 3/04 G16H 70/60
Abstract:
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.
Dr. Gao graduated from the Beijing Med Univ, Beijing City, Beijing, China in 1986. She works in Carson City, NV and 1 other location and specializes in Gastroenterology and Internal Medicine. Dr. Gao is affiliated with Carson Tahoe Regional Medical Center.