Method to Predict, Evaluate and Map Soil Microbial Function for Broad Geographical Regions using Omics and Machine Learning

Case ID:

This invention is a method for predicting and geographically mapping the functional composition of soil microbial communities across a large geographical area and with diverse environments using machine learning. The resulting information is relevant to evaluating soil carbon, nitrogen, and phosphorus cycling.

Microbial-mediated soil organic matter decomposition regulates many key ecosystem functions such as soil nutrient cycling, carbon sequestration, and soil fertility. However, representing microbial processes in Earth system models is still a challenge due to the lack of a clear understanding regarding the spatial patterns of functional diversity within microbial communities and how diverse environments regulate these communities.


  • Biomass composition for food/fuels/chemicals
  • Farm/pasture/forest management
  • Other land management practices
  • Consulting for policy evaluation


  • More efficient
  • Unique predictive technology
  • Evaluates multiple elements simultaneously
Patent Information:
Contact For More Information:
Jonathan Larson
Senior Licensing Manager, College of Science
The University of Arizona
Lead Inventor(s):
Yang Song
Changpeng Fan