Back to results
US 11,028,449 B2Granted
Microbiome based systems, apparatus and methods for monitoring and controlling industrial processes and systems
AssigneeBiota Technology, Inc.
InventorsRob Knight, Ajay Kshatriya, Chris Lauber, J. Gregory Caporaso, Dan Knights, Ryan Gill, Joel Moxley
Life of this patent
Estimating the term…
Abstract
There are provided methods, systems and processes for the utilization of microbial and related genetic information for use in industrial settings, such as the exploration, determination, and recovery of natural resources, minerals, and energy sources, the monitoring and analysis of processes, activities, and materials transmission.
Claims20
A computerized method comprising:
- generating, using one or more computer processors, microbiome information from genetic material extracted from a first sample of a plurality samples sourced from a plurality of wells at a location associated with an industrial operation;
- generating, using the one or more computer processors, microbiome information from genetic material extracted from a second sample of the plurality of samples;
- selecting, using the one or more computer processors, a machine-learned model from a plurality of machine-learned models based on a testing of a predictive accuracy of the machine-learned model when applied to a plurality of subsets of the microbiome information from the genetic material extracted from the first sample, the selecting based on a class of the machine-learned model being a specialized generative model that was generated using an embedded approach to feature selection having knowledge of a parameter-selection process;
- generating, using the one or more computer processors, predictive microbiome information corresponding to the microbiome information from the genetic material extracted from the second sample, the generating of the predictive microbiome information including applying the selected machine-learned model to the genetic material from the second sample, the predictive microbiome information relating to a source production zone corresponding to a hydrocarbon formation or a characteristic of a reservoir corresponding to the hydrocarbon formation, the characteristic including at least one of pressure, temperature, or viscosity;
- and
- communicating, using the one or more computer processors, the predictive microbiome information relating to the source production zone or the characteristic of the reservoir for display in a user interface.
3 more elements
A method comprising:
- selecting, using one or more computer processors, a machine-learned model from a plurality of machine-learned models based on a testing of a predictive accuracy of the machine-learned model when applied to a plurality of subsets of microbiome information from a genetic material extracted from a first sample, the selecting based on a class of the machine-learned model being a specialized generative model that was generated using an embedded approach to feature selection having knowledge of a parameter-selection process;
- generating, using one or more computer processors, predictive microbiome information based on a nucleic acid sequencing data, the generating of the predictive microbiome information including applying a machine-learned model to the nucleic acid sequencing data, the predictive microbiome information predicting a productivity of each of a plurality of wells at a location based on intra-well temporal dynamics of microbial communities at the plurality of wells;
- communicating, using the one or more computer processors, the predictive microbiome information for display in a user interface.
A method comprising:
- performing operations at a computing system having one or more computer processors for communicating predictive microbiome information to a memory storage device for display in a human machine interface (HMI) of the memory storage device, the display having an interactive configuration for identifying patterns in the predictive microbiome information that are relevant to optimizing recovery of hydrocarbons, the operations comprising:
- selecting, using the one or more computer processors, a machine-learned model from a plurality of machine-learned models based on a testing of a predictive accuracy of the machine-learned model when applied to a plurality of subsets of the microbiome information from a genetic material extracted from the first sample, the selecting based on a class of the machine-learned model being a specialized generative model that was generated using an embedded approach to feature selection having knowledge of a parameter-selection process;
- generating, using the one or more computer processors, the predictive microbiome information, the predictive microbiome information corresponding to sequencing data extracted from genetic material from a second sample sourced at a well or an additional well, the generating of the predictive microbiome information including applying the selected machine-learned model to the sequencing data extracted from the genetic material from the second sample, the predictive microbiome information relating to a source production zone corresponding to a hydrocarbon formation or a characteristic of a reservoir corresponding to the hydrocarbon formation, the characteristic including at least one of pressure, temperature, or viscosity;
- and
- performing, using the one or more computer processors, the communicating of the predictive microbiome information for the display in a user interface.
2 more elements
Dependencies are inferred from the claim text.