Viruses

Researchers at MIT and elsewhere developed an artificial intelligence predictive model that can be used to detect which strains of Covid-19 could become dominant and lead to a new wave of illness, reports Ruth Reader, Carmen Paun, Daniel Payne and Eric Schumaker for Politico. The model, “found three strong predictors of a dominant variant: the number of infections a strain causes in its first week relative to the number of times it appears in sequencing, the number of mutations in the spike protein, and the number of weeks since the current dominant variant began circulating,” they note.

Prof. Rudolf Jaenisch speaks with Fortune reporter Carolyn Barber about the immune system’s response to Covid-19. “The immune system is really stimulated–it’s in hyperdrive,” says Jaenisch. 

Prof. Kevin Esvelt and his students have found that language-generating AI models could make it easier to create pandemic potential pathogens, reports Kelsey Piper for Vox.

MIT researchers and an undergraduate class found that chatbots could be prompted to suggest pandemic pathogens, including specific information not commonly known among experts, reports Ryan Health for Axios. The MIT researchers recommend "pre-release evaluations of LLMs by third parties, curating training datasets to remove harmful concepts, and verifiably screening all DNA generated by synthesis providers or used by contract research organizations."

Science reporter Robert F. Service spotlights how Prof. Kevin Esvelt is sounding the alarm that “AI could help somebody with no science background and evil intentions design and order a virus capable of unleashing a pandemic.” 

Using an artificial intelligence system, researchers at MIT and elsewhere have developed a new Covid-19 vaccine that could be effective against current and future strains, reports Gwen Egan for Boston.com. “The vaccine differs from others currently on the market due to the portion of the virus being targeted,” writes Egan.  

Boston Globe reporter Hiawatha Bray writes that MIT researchers have used an AI system to identify a potential new Covid-19 vaccine that may be effective against both current and future variants of the virus. “The new vaccine targets a portion of the COVID virus that is much less prone to evolve,” writes Bray. “That could potentially make it effective against many different versions of the virus, eliminating the need for routine booster shots.”

MIT researchers have developed a new approach to vaccines that uses “a machine learning twist [that] could put an end to boosters and seasonal variant shots,” reports Devin Coldewey for TechCrunch.

Vox reporter Kelsey Piper writes that a new report by Prof. Kevin Esvelt provides a roadmap for how to prepare for the next pandemic. In the report, Esvelt emphasizes that: “We’re not helpless, whether against nature or malign actions by human beings. We do have to invest in actually being prepared, but if we’re prepared, we could weather even a worst-case scenario: a deliberate release of a human-made virus engineered to be both extra deadly and extra contagious.”

Newsha Ghaeli ’17 - president and co-founder of Biobot, a public health research, data and analytics firms that has developed and promoted wastewater surveillance technology - speaks with Vox reporter Muizz Akhtar about how wastewater surveillance can be used to predict and prepare for future pandemics. “Our vision is that this is a permanent infrastructure layer on our sewer systems, so that it becomes one of the core kinds of pandemic preparedness in this country and disease surveillance globally,” says Ghaeli.

Boston Globe Magazine reporter Courtney Humphries spotlights MIT startup Biobot Analytics, co-founded by Mariana Matus ’18 and Newsha Ghaeli ’17, for using their wastewater and sewage tracking technology to identify Covid -19 in communities across the United States. “Because people shed the virus in their stool before they test positive, Biobot’s data are often a leading indicator of where the infection is heading, arriving ahead of case counts by a few days,” writes Humphries.

Boston Globe reporter Pranshu Verma spotlights MIT startup Biobot Analytics, co-founded by Mariana Matus ’18 and Newsha Ghaeli ’17, for their work studying sewage data to better predict the spread of Covid-19 in communities. “For health officials, it [the data] confirms whether Covid spikes in the community are real, and not due to increased testing or other factors,” writes Verma. “Moreover, Covid levels in waste water are a leading indicator for new clinical cases, giving health officials a few days’ notice if they’ll see more sick patients showing symptoms.”

MIT researchers are developing innovations aimed at improving Covid-19 diagnostics, including an atomic-level test designed to increase testing accuracy, reports Steven Zeitchik for The Washington Post. Professor James Collins and his team are developing “a mask that uses freeze-dried technology to detect the coronavirus.”

Boston Globe reporter Kay Lazar spotlights how the Broad Institute “has become the region’s powerhouse for monitoring shifts in the genetic makeup of the coronavirus.”

Writing for The Washington Post, Prof. Kevin Esvelt argues that research aimed at creating pandemic-causing viruses should be considered a matter of international security. “Natural pandemics may be inevitable. Synthetic ones, constructed with full knowledge of society’s vulnerabilities, are not,” writes Esvelt. “Let’s not learn to make pandemics until we can reliably defend against them.”