ByCRAIG VENTER

AI-can-dectec-disease-earliest-stage
Combining genomics, MRI scans and artificial intelligence will usher in a new era in healthcare

In many countries, Covid-19 has spread because of a popular scepticism about science, a political manipulation of data and an abundance of inaccurate information, spread in large part on social media but fuelled at some of the highest levels of governments. In 2021, we will understand that only by developing new, science-based approaches to disease detection will we avoid similar future catastrophes.

My own country, the United States, has already provided a live demonstration of this fact. States that enforced practices such as social isolation/distancing, hand washing and use of sanitisers and face masks have had the lowest per capita rates of Covid-19 infection, averaging around 100 to 200 cases per 100,000 people. Compare that figure to the 2,300 per 100,000 people in states that did not enforce these measures.

Reconnecting science with healthcare will have impressive results. One weakness of the world’s response to the pandemic has been its unwillingness to use science to predict vulnerabilities in individuals before they become ill. Not only have many countries been reluctant to perform widespread testing on people who show no Covid-19 symptoms, but they have also been caught seemingly by surprise when pre-existing conditions have exacerbated the disease. In 2021, we will see the application of a multimodal testing approach to detect propensities to diseases in people at very early stages. 

We are used to the idea that sequencing an individual’s genome can give a reasonable indication of the likelihood of developing a particular disease. In 2021, we will combine genomics with MRI imaging and machine learning to provide a more sophisticated snapshot of an individual’s health. In research published in 2020, my colleagues at Human Longevity and the J Craig Venter Institute, of both of which I am the founder, analysed data collected on thousands of individuals. By integrating whole-genome sequencing with advanced imaging and the analysis of blood metabolites, they were able to identify adults at risk for key conditions including cancer, heart disease, diabetes, chronic liver disease and more. 

The study demonstrated the importance of multimodal testing, rather than relying solely on the sequencing of an individual’s genotype. A lack of phenotype and genotype associations were observed in 5.8 per cent of individuals with pathogenic genetic variants, further suggesting that the identification of pathogenic genetic variants by sequencing alone is not sufficient for a definitive diagnosis.

We intend to scale up this study in 2021 to test multimodal, preventative-medicine procedures on a larger cohort. The pandemic has temporarily derailed this effort but, with strong, knowledge-based leadership and a renewed trust in science, my hope is that 2021 will see us taking a significant leap towards a data-driven, multimodal approach to detecting disease.

Craig Venter PhD is founder and CEO of the J Craig Venter Institute

Via Wired.com