When I started suing food companies more than 30 years ago, after E. coli O157:H7 in undercooked Jack in the Box hamburgers killed four children and sickened hundreds, finding out where the pathogen came from proved slow, difficult, and often impossible. Investigators matched the bacteria using a crude “fingerprinting” method that made two samples look similar but gave little indication of whether they were actually the same organism.
That world is gone. Whole genome sequencing, which reads a bacterium’s complete DNA, has been done for foodborne illnesses, much as DNA testing has been done for forensics. Today, when Listeria shows up in a California nursing home resident and a frozen nutritional shake produced 2,000 miles away, public health scientists can say with almost certainty that it is the same organism. The FDA’s GenomeTrakr and CDC’s PulseNet databases and NCBI’s website currently hold hundreds of thousands of genomic fingerprints. The 2025 Lyon Magnus outbreak (42 infections and 14 deaths in 21 cantons) was addressed in this way. So did the largest and deadliest listeriosis outbreak in history. The 2017-2018 disaster in South Africa, which killed 216 people, can be traced back to Poloni at the Tiger Brands plant in Polokwane through a single sequence type called ST6.
When it comes to public health, the case is overwhelming. WGS reduces the scale of outbreaks by finding the source faster, narrows the scope of recalls to target problematic products instead of the entire product, and captures contamination at near-low levels. None of that is seriously controversial.
The harder question, and the one companies actually ask me, is whether food producers should implement the same technology within their own products and factories.
Here’s the uncomfortable part. Let me be clear, because industry lawyers generally don’t do that. Positive results obtained today can be used against you tomorrow. If you swab a plant, find listeria in a floor drain, sequence it, and later find its genome matches that of a sick child, you’re tied to the case in a way a jury won’t forget. Some defense attorneys take the logical next step and quietly advise their clients to do less testing. Because what you don’t know can’t be summoned.
I’ve spent decades on the wrong side of that advice, and I’m here to tell you that bad advice is morally bankrupt and, in the long run, a legal disaster.
Choosing not to see is not a defense. This is willful blindness, and the jury understands it immediately. Companies that find, document, and surgically remove resident strains have a concurrent track record of doing exactly what a responsible food producer should do. The company decided that ignorance was safer and kept no records. Since there was no outbreak until a larger, deadlier outbreak occurred, plaintiff lawyers like me ask under oath why a company with hundreds of millions of dollars in revenue would choose not to use technology that public health laboratories in developing countries could afford.
And WGS works in both directions. The same accuracy that can implicate you can also get you clean. When your product is swept up in an unrelated epidemic, your genomic data is the fastest route to proving that your disease is not a disease. That means getting your name in the headlines and getting your truck back on the road. I watched the sequence prove the innocence of the accused creators as emphatically as it did their convictions.
WGS is not magic. This requires real-world bioinformatics expertise, validated pipelines, and reference databases for which there are still gaps for rare and emerging pathogens.
But none of this is an argument for staying in the dark. It has been reflexive for the food industry to know as little as possible about its own contamination, on the theory that what you don’t know can’t hurt you in court. Genome has closed that deal. Plant pathogens already have fingerprints. The only question is whether you’ll be willing to read it before someone else’s child does.
