Google Inc.

Exelixis, Inc.

Amazon Technologies, Inc.

Exelixis Pharmaceuticals, Inc.

Bindu Priya Reddy

Andrew Roy Buchman

Ning Mosberger

Gaurav Bhaya

David Andrew Ruddy

Helen Louise Francis-Lang

Sarah Sirajuddin

Jennifer L Kozenski

Arvind Sundararajan

David Kale

Jonathan Scott Margolis

Puneet Agarwal

Stefano Stefani

Stephen Kohl Doberstein

John William Winslow

Lawrence Jonathan Brunsman

Madan Mohan Rao Jampani

Darren Platt

Jonathan Brunsman

Yuling Luo

Candace Swimmer

Sheila Akiko Homburger

Casey Kopczynski

Jonathan Cagampang Heller

Allen James Ebens, Jr

Marshall Spight

Hyunjoon Song

Shuyi Zhang

Venkatesh Maralavadi Sreenivas

Catherine Sue Erickson

Christian Burks

Lucile A Gillett

Linda Nolan Keyes

Gaurav Ravindra Bhaya

Systems and techniques are disclosed for a centralized platform for enhanced automated machine learning using disparate datasets. An example method includes receiving user specification of one or more data sources to be integrated with the system, the data sources storing datasets to be utilized to train one or more machine learning models by the system, and the datasets reflecting user interaction data. A dataset is imported from the data source, and machine learning models are automatically trained based a particular machine learning model recipe of a plurality of machine learning model recipes. A first trained machine learning model is implemented, with the system being configured to respond to queries based on the implemented machine learning model, and with the responses including personalized recommendations.

Centralized platform for enhanced automated machine learning using disparate datasets

A method and system associating labels and attribute values with items in a collection of data. Providers can associate attributes and labels with their data or attributes and labels can be added to existing data. A preferred embodiment allows a content provider to upload data and to attach their own custom labels and attributes to items or to use predefined labels and attributes. Providers can upload data using a user interface or a bulk upload mechanism.

Adding attributes and labels to structured data

A user can refine a search over structured data by specifying that a label or an attribute value be used to further filter the results of a query.

Search over structured data

The present invention relates to Drosophila genes and methods for their use. The invention provides nucleotide sequences of Drosophila genes, amino acid sequences of the encoded proteins, and derivatives (e.g., fragments) and analogs thereof. The invention further relates to fragments (and derivatives and analogs thereof) of proteins which comprise one or more domains of a Drosophila protein. Antibodies to Drosophila proteins, and derivatives and analogs thereof, are also provided. Also provided herein are vectors and host cells comprising such nucleic acids. Methods of production of a Drosophila protein (e.g., by recombinant means), and derivatives and analogs thereof, are provided. Chimeric polypeptide molecules comprising polypeptides of the invention fused to heterologous polypeptide sequences are provided. Methods to identify the biological function of a Drosophila gene are provided, including various methods for the functional modification (e.g., overexpression, underexpression, mutation, knock-out) of one gene, or of two or more genes simultaneously. Methods to identify a Drosophila gene which modifies the function of, and/or functions in a downstream pathway from, another gene are provided. The invention further provides for use of Drosophila proteins as media additives or pesticides.

Drosophila sequences

Tandem pore domain weak inward rectifying K (TWIK) channel nucleic acids and proteins that have been isolated from and Leptinotarsa are described. The TWIK channel nucleic acids and proteins can be used to genetically modify metazoan invertebrate organisms, such as insects, coelomates, and pseudocoelomates, or cultured cells, resulting in TWIK channel expression or mis-expression. The genetically modified organisms or cells can be used in screening assays to identify candidate compounds which are potential pesticidal agents or therapeutics that interact with TWIK channel proteins. They can also be used in methods for studying TWIK channel activity and identifying other genes that modulate the function of, or interact with, the TWIK channel gene.

Nucleic acids and polypeptides of invertebrate TWIK channels and methods of use

The present invention relates to insulin-like genes and methods for identifying insulin-like genes. The methods provide nucleotide sequences of insulin-like genes, amino acid sequences of their encoded proteins, and derivatives (e.g., fragments) and analogs thereof. The invention further relates to fragments (and derivatives and analogs thereof) of insulin-like proteins which comprise one or more domains of an insulin-like protein. Antibodies to an insulin-like protein, and derivatives and analogs thereof, are provided. Methods of production of an insulin-like protein (e.g., by recombinant means), and derivatives and analogs thereof, are provided. Further, methods to identify the biological function of a insulin-like gene are provided, including various methods for the functional modification (e.g., overexpression, underexpression, mutation, knock-out) of one or more genes simultaneously. Still further, methods to identify a gene which modifies the function of, and/or functions in a signaling pathway with an insulin-like gene are provided. The invention further provides uses of Drosophila insulin-like nucliec acids and proteins, e.g., as media additives, and as pesticides.

Nucleic acids and proteins of D. melanogaster insulin-like genes and uses thereof

Growing community of inventors

Foster City, CA, United States of America

Bindu Priya Reddy