Using insects to understand and protect biodiversity
Project Premonition began as an effort to use AI to detect and track emerging infectious diseases. As an AI for Earth project, researchers are applying the same technology to an issue perhaps just as complicated as public health: biodiversity.
Conservationists need fast, reliable data on the biodiversity in a given area, but gathering samples in traditional ways is tedious and time-consuming. Project Premonition uses mosquitoes as field biologists, analyzing the blood that mosquitoes take from animals in the environment to glean valuable data. To achieve this, we’re developing drones that autonomously locate mosquito hotspots, robotic traps to collect specimens, and cloud-scale genomics and machine learning algorithms to identify each animal bitten.
These advances make it possible to collect species data faster, process genetic material more efficiently, and quickly derive insights that can help conservationists protect and conserve biodiversity.
Robotic insect traps, paired with cloud-scale genomics and machine learning algorithms, enable biodiversity insights
1. HOST SPECIES
The species of focus and the type of insect that feeds on that species are identified.
Insects travel through different populations to feed. The blood meal they take from each bite carries the animal’s DNA.
3. PROJECT PREMONITION SMART TRAPS
These intelligent robots sort insects by analyzing wingbeat frequency to differentiate the species, preserving those that feed on the animal species being studied.
The host blood is extracted from the insects and the host species DNA is sequenced.
Genetic info is analyzed to identify biodiversity and estimate an ecosystem's health.
Researchers can create a heat map detailing population density throughout the tested area. This can be used to track population growth as well as movement over time.