Nature Biotechnology features Debug Fresno
Monday, April 6, 2020
Today, our scientific paper reporting near-elimination of Aedes aegypti from three neighborhoods in Fresno County, California was published in the journal Nature Biotechnology. While we previously summarized the impressive topline results of our Debug Fresno program, the peer-reviewed paper provides extensive new details regarding our field study design, mosquito rearing and release technology, trapping data, and analysis methods. The general public, mosquito control community, governmental stakeholders, and regulators alike can now independently review our methods and results.
Debug-built sex sorters achieve unprecedented accuracy by classifying individual males and females using industrial vision and machine learning algorithms.
Of particular interest, the manuscript reveals formerly undisclosed features and performance metrics of our automated mosquito production and release tools. In the paper, we describe how we reared mosquito larvae in disposable, thermoformed plastic containers, resulting in unprecedented consistency in male mosquito production. We also provide a step-by-step breakdown of our automated sex separation process including calculations of accuracy at each step. For the end-to-end process, we achieved a remarkable level of accuracy that is orders of magnitude better than manual sex separation methods. On the release side, we provide visualizations of how custom software guides male mosquito releases from our vans and allows us to detect and target those areas where releasing additional male mosquitoes would further improve program effectiveness. In addition to these metrics and descriptions, we are releasing all of the field data we collected during the course of the study to allow for analytical replication.
Debug Fresno was an incredibly rewarding partnership among ourselves, the Consolidated Mosquito Abatement District, MosquitoMate, and the residents of Fresno County. As our first field trial, it taught us many lessons about how to successfully run a sterile insect technique (SIT) program. Publication of the results from this field study marks a critical point in the evolution of the Debug project. We now have clear, peer-reviewed evidence that Debug technology can be used to massively suppress the number of biting female Aedes aegypti by automating both mass-rearing and release of sterile male mosquitoes. More than 2 billion people are at risk of contracting Aedes-borne viruses such as dengue, Zika, and chikungunya, and the global distribution of Aedes aegypti continues to expand worldwide. Our results give hope that we—in collaboration with forward-thinking partners—can turn the tide in this ongoing global health challenge. We have already incorporated what we learned from this trial into our other programs around the world, including our work with The National Environment Agency (NEA) in Singapore, and look forward to expanding to additional communities that routinely suffer from outbreaks of Aedes-transmitted diseases.
Jacob Crawford, PhD, Sr. Scientist, Verily and Brad White, PhD, Debug Lead Scientist, Verily
Debug-built sex sorters achieve unprecedented accuracy by classifying individual males and females using industrial vision and machine learning algorithms.
Of particular interest, the manuscript reveals formerly undisclosed features and performance metrics of our automated mosquito production and release tools. In the paper, we describe how we reared mosquito larvae in disposable, thermoformed plastic containers, resulting in unprecedented consistency in male mosquito production. We also provide a step-by-step breakdown of our automated sex separation process including calculations of accuracy at each step. For the end-to-end process, we achieved a remarkable level of accuracy that is orders of magnitude better than manual sex separation methods. On the release side, we provide visualizations of how custom software guides male mosquito releases from our vans and allows us to detect and target those areas where releasing additional male mosquitoes would further improve program effectiveness. In addition to these metrics and descriptions, we are releasing all of the field data we collected during the course of the study to allow for analytical replication.
Debug Fresno was an incredibly rewarding partnership among ourselves, the Consolidated Mosquito Abatement District, MosquitoMate, and the residents of Fresno County. As our first field trial, it taught us many lessons about how to successfully run a sterile insect technique (SIT) program. Publication of the results from this field study marks a critical point in the evolution of the Debug project. We now have clear, peer-reviewed evidence that Debug technology can be used to massively suppress the number of biting female Aedes aegypti by automating both mass-rearing and release of sterile male mosquitoes. More than 2 billion people are at risk of contracting Aedes-borne viruses such as dengue, Zika, and chikungunya, and the global distribution of Aedes aegypti continues to expand worldwide. Our results give hope that we—in collaboration with forward-thinking partners—can turn the tide in this ongoing global health challenge. We have already incorporated what we learned from this trial into our other programs around the world, including our work with The National Environment Agency (NEA) in Singapore, and look forward to expanding to additional communities that routinely suffer from outbreaks of Aedes-transmitted diseases.
Jacob Crawford, PhD, Sr. Scientist, Verily and Brad White, PhD, Debug Lead Scientist, Verily