Auto-omics for climate and sustainability

An automated explainable bioinformatics and AI workflow for multi-omic, climate and environmental data, applied to sustainability problems e.g., nature-based carbon capture.


As climate change becomes ever-more urgent, we’re investigating nature-based processes for carbon capture from the atmosphere and how to trap it in biomass, soil, deep oceans, or underground reservoirs. Nature is very efficient at converting CO2 to biomass through photosynthesis by plants and by CO2 mineralization (to carbonates) through microbial activities in soil. In fact, extracting CO2 from the atmosphere using natural processes is the most effective and fastest implementable approach that we currently know. AI driven analytics can optimize strategies to enhance natural based carbon sequestration and enable more accurate means to measure and verify the carbon stored in soil and plants.


We’redeveloping computational methods and technologies to determine which genetic attributes help organisms capture carbon and adapt to specific environmental conditions caused by climate change. This will enable our partners to introduce identified genetic biodiversity in the form of advantageous plant traits or soil microorganisms for carbon capture. Our AI framework enables the integration of genetic information (profiled by multi-omics technologies) with representations of the environmental conditions for an organism. Environmental data is sourced using the IBM Environmental Intelligence Suite. Omic data encompasses sequencing technologies that characterize genetic variation, profiling, genomics (DNA), transcriptomics (RNA), metagenomics (DNA of the microbiome), proteomics (proteins), and epigenomics (chemical modifications to DNA). This project takes a data-driven approach to develop the methodological components of a large-scale geospatial AI framework that integrates multi-omic datasets with environmental, geospatial, and temporal data to accelerate discovery for environmental genetics. Using explainable AI, we can also identify integrated fingerprints of genetic and environmental features that act as biomarkers for carbon capture.

Some of our current projects:

  • Developing bioinformatics and graph-based AI methods to allow representation of the soil microbiome (via metagenomic and environmental soil and climate geospatial data) and to predict its potential for carbon sequestration, with a goal of enabling manipulation or optimization through things like smart fertilizers.

  • Developing a novel pan-genome bioinformatics and machine-learning solution to predict key traits of interest while also identifying predictive genetic variants that might improve the efficiency of photosynthesis in plants.

  • Developing capability for time-series transcriptomic bioinformatics and ML analysis to ascertain the impact of various treatments, such as fertilizers, on plant efficiency and soil health.




Related projects

Host-pathogen interactions.png

Host-pathogen interactions for healthcare and drug discovery

An automated explainable bioinformatics and deep learning workflow to elucidate microbiome-metabolite relationships during a host-pathogen interaction.

Auto-omics for healthcare and drug discovery

An automated explainable bioinformatics and AI workflow for multi-omic, clinical and experimental data, applied to healthcare and drug discovery problems.