A comprehensive new gene cataloging project, says IBM research scientist James Kaufman, could help food producers prevent contamination on a massive scale.
James H. Kaufman believes an ounce of prevention is worth a pound of cure when it comes to food safety. That’s why he’s excited to be one of the leaders of an ambitious new project to catalog all the active genes that typically exist in different food environments, from farms through transport, processing and distribution—under normal conditions and when contamination occurs.
The resulting wealth of new data, says the IBM computational biologist, will revolutionize the safety of our food supply. Today food safety is about detecting pathogens—ideally during production, but all too often in foods that have already gone to market and infected people. Kaufman and his colleagues want to reduce foodborne illness by making the system more predictive. If suppliers and manufacturers can detect changes in the life forms present in all foods that could lead to contamination, they will be better able to prevent the contamination from happening.
Every food is a habitat, and if you disturb it, its microbial community will respond.”
Announced earlier in 2015 by joint sponsors IBM and Mars, the project is called the Consortium for Sequencing the Food Supply Chain. Instead of just isolating known pathogens such as salmonella and listeria, the initiative will use metagenomics—analyzing microbial genetic material collected from environmental samples, such as swabs from storage bins or processing vats in a food manufacturing or distribution facility.
“Think of an ingredient as an ecosystem. There are communities of living organisms in all foods. We want to know what goes on during their normal metabolism,” says Kaufman, manager of public health research at IBM’s Almaden research center in San Jose, Calif. “If you understand what that community looks like under normal conditions, then changes in those conditions can alert you that something may be wrong.”
Putting the puzzle together
Kaufman, who holds a doctorate in physics from the University of California at Santa Barbara, has worked on topics ranging from electronic health records to disease surveillance. But even for a researcher who enjoys “solving hard problems with computers,” analyzing the metagenomics of the food supply chain is a daunting mission. The project will sequence all of the DNA and RNA (ribonucleic acid) that scientists find in environmental samples, first under normal conditions and then spiked with heavy metals or chemicals to see the biological response.
Kaufman says metagenomics offers an opportunity to create data management systems that will help food companies and regulators organize enormous quantities of information on the microbial ecology of common food ingredients. IBM is building a cloud computing system to manage the data, and will develop algorithms and modeling tools that will make it possible to sort through the information and look for patterns and anomalies.
“Genomic data is like little puzzle pieces, so you need computing to put the pieces together. This project is noncompetitive, but hopefully it will become a business opportunity down the road,” Kaufman says.
This isn’t the first time IBM and Mars have worked together on food genetics. The companies teamed up in 2008 with the U.S. Department of Agriculture and academic researchers to sequence the genome of cacao (cocoa), a major agricultural commodity and the basic ingredient in chocolate. Mars, which manufactures Snickers, M&Ms and many other popular candy brands, was interested in identifying genes associated with flavorful and hardy cacao cultivars. Genetic maps and other information from the project, which enable farmers to plant stronger and more pest-resistant cocoa, are available online.
The Consortium for Sequencing the Food Supply Chain grew out of conversations between Mars and IBM executives at a 2014 conference on potential applications from metagenomics. “Mars has operations worldwide and is a leader on food safety, but they asked a really important question: How could they test for threats that weren’t known yet?” says Kaufman. “We saw that metagenomics could revolutionize food safety.”
The first phase of work has already started at a Mars pet food factory near Reno, Nev. Researchers will sample and sequence a half-dozen major ingredients, such as chicken meal and cornmeal, when they arrive at the plant and as they move through it, analyzing how the microbial communities within the foods vary seasonally and as they are processed. Periodically, samples will be spiked with heavy metals or other contaminants to see how the foods’ microbiomes respond.
“We are using several state-of-the-art techniques, including next-generation sequencing [an approach that sequences millions of small DNA fragments in parallel],” says Kaufman. “And we plan to compare the metagenomic data that we produce using new techniques with the results of standard approaches that are in common use today.”
The analysis will take place at IBM Almaden and in microbiology laboratories at collaborating universities. Kaufman estimates about two years will be needed to produce metagenomic sequences for the target ingredients at the Reno factory. Results will be published in peer-reviewed journals and later made publicly available for free.
A wide variety of potential benefits
As the food industry becomes increasingly globalized, Mars and other companies that import ingredients from abroad have clear interests in ensuring their supply chains are safe. The Centers for Disease Control and Prevention (CDC) has warned in recent years that a growing share of foodborne illness outbreaks are being caused by imported foods, including seafood, spices and fresh produce. Information on the microbial environment of common foods could lead to the development of new fast screening tests for previously unknown pathogens and contaminants. With this kind of data, inspectors will be better able to catch problems in imported ingredients before they are blended into products for the U.S. market.
Kaufman also expects that data from the sequencing project will help food producers and processors manage perishability. “We see lots of opportunities to reduce food waste by identifying genes that signal spoilage,” he says. “For example, the produce industry is very interested in moving from a first-in-first-out delivery model to a first-to-expire model.” With a better understanding of the microbiomes of their products, suppliers will know which foods should be shipped first to supermarkets. Minimizing spoilage at every level of the produce supply chain will mean that fewer fruits and vegetables are thrown away and ensure that consumers receive better-quality products.
The new data also could suggest new ways to retard spoilage. Some foods contain organisms in their natural microbiomes that inhibit the growth of pathogens like salmonella and staphylococcus aureus. “You can imagine developing probiotics based on these for a type of food,” Kaufman says. “People have been adding helpful organisms to food to retard spoilage for centuries.” Fermented foods such as wine, cheese and yogurt, for example, use beneficial organisms in their production.
If the consortium grows as planned, Kaufman expects that within five years it will have produced a large body of data on genes that should be present in foods and on other genes that signal spoilage. “Every food is a habitat, and if you disturb it, its microbial community will respond,” he says. “We want the things we learn to be of value up and down the supply chain, and we would like to teach what we learn through this project to suppliers around the world.”