What does big data have to do with food safety?
A lot, it turns out.
Some food safety researchers mine Yelp or other social media to detect clusters of illnesses (a bunch of people posting negative reviews on a restaurant around the same time referencing illness is a good clue to look closer). Unfortunately, I was unable to find an image of this in action. I think they use a software designed to flag out clusters of illness; I doubt someone sits at a desk eating popcorn while reading angry Yelp reviews, as much fun as that might be
Some researchers compare massive datasets of entire organisms DNA with each other, looking for clues on how bacteria adapt to different conditions, or to determine which Salmonella are part of an outbreak vs background noise. Then they turn the data into big trees which tell them how all the Salmonella in their dataset are related to each other. This research typically involves a supercomputer (which unfortunately isn’t quite as cool as it sounds).
Some researchers (myself included), look at giant spreadsheets of information on which genes a bacterium uses to survive heat, or form a biofilm (see below). It’s not sexy, but it’s important. It’s funny that something as low tech as an Excel spreadsheet can hold vital clues for science. After all, it’s literally a list of cryptic names (for bits of DNA/RNA) and numbers (how much is that DNA used when the bacteria is living in condition X?)
The people who best combine big data and food safety are the microbiology-trained coders. Not only can these elusive creatures write software programs to compare bacterial DNA, they can also get why a program for bacterial DNA might need to work a little differently than one for human DNA. They are wonderful, and we all long to have one as a friend. Honestly, though, these guys are so rare they might as well be unicorns. Or coelocanths. But they can get a job anywhere(hint: good career option).
Ok. I need to get back to making sense of my own big datasets… More another time