After Covid, PCR Tests Are Coming For Cotton

KEY INSIGHTS

• Companies such as Applied DNA Sciences and Oritain can identify the source of a cotton sample by analysing its genetic or chemical fingerprint.
• The technology is in demand as brands and retailers face mounting pressure to prove sustainability claims or assure regulators they aren’t using cotton from regions where forced labour is a risk.
• But to make this sort of testing fast, cheap and widespread will require overcoming some major obstacles.

It’s a simple enough idea in theory: fashion brands worried about the origins of their cotton could put a fibre sample in a machine, run a quick test and know precisely what area it came from.

The technology already exists and is in demand as brands face pressure to rid their clothes of cotton harvested with forced labour or falsely labelled as GMO-free. Polymerase chain reaction, or PCR, testing, deploys the same method used to detect Covid to analyze cotton’s DNA. The results can then be checked against a database of known samples to separate, say, organic material grown in Gujarat from cotton coming from Xinjiang, which the US banned from import last year.

But turning the idea into reality isn’t so simple.

While testing a single type of cotton is straightforward enough, the process becomes trickier when a fabric is made with cotton from multiple sources or blended with synthetics. Building a comprehensive database of every known type of cotton also requires groundwork no single testing company has managed to date.

“If you could do a PCR test for Covid, then why couldn’t you do PCR tests for cotton?” said MeiLin Wan, vice president of textiles at Applied DNA Sciences, which does cotton DNA testing and also makes PCR tests for Covid-19. “It’s all a question of will and intention and resources.”

If you could do a PCR test for Covid, then why couldn’t you do PCR tests for cotton?

Another promising method, isotopic analysis, looks at a fibre’s chemical fingerprint, which will vary according to the environment the cotton grew in based on factors such as the altitude, rainfall and composition of the soil. It’s also a powerful technology — the same law enforcement may use to determine where drugs, or a body, came from — but it can run into similar challenges.

Proponents of both technologies say the industry is likely years away from this type of testing becoming fast, cheap and widespread. Still, the promise of skipping to the beginning of a complex international supply chain to weed out problematic supplies is seductive.

“It can be a good supplement,” said Nate Herman, senior vice president of policy at the American Apparel and Footwear Association. “It’s not the silver bullet that some people are hoping for.”

The Promise and the Challenge

Applied DNA operates on the premise that cotton grown in different areas has different genetic markers, which can be used to decipher where a sample came from. Wan said the technology has been around for more than a decade and is extremely reliable, if not quite as quick or inexpensive as a PCR test for Covid.

But decoding the genome of a cotton sample isn’t of much use on its own. To know if it came from Texas or Xinjiang, you need cotton from each place to compare it against.

“We don’t have an exhaustive library of every single cotton type and that’s the problem with finished products,” Wan said. “They could’ve used five different cottons as well in that sample, which is highly conceivable.”

Applied DNA focuses on confirming what brands should already know about their cotton — for instance, that cotton labelled Egyptian and GMO-free is in fact from Egypt and free of GMOs. A 2016 scandal involving falsely labelled Egyptian cotton forced several large US retailers to pull bedding from their shelves. The company has collected samples from big cotton-producing countries such as India and China to fill out its database but is far from a complete record.

Isotopic analysis, which Applied DNA and others can apply to cotton, offers another tracing option, but it has its own shortcomings.

“That’s something where you have to be able to constantly refresh the samples,” said Herman. “If, because of climate change, that area has experienced a drought for two years, that’s going to change the chemical makeup of a raw material.”

Oritain, which combines isotopic analysis with data science, works with partners such as Kering and Primark on tracing programs. Grant Cochrane, its chief executive, said the company developed its algorithms to counter the difficulty of analysing blended cotton, and that it has ways to monitor and predict changes resulting from shifts in weather. Its data library reaches back more than a decade.

It claims to have mapped 90 percent of the world’s cotton-growing regions, which isn’t just useful for matching a sample to a known location. If you get a fibre and can’t identify the origin, you know where it’s not coming from at least.

Cochrane wouldn’t elaborate on whether Oritain has samples from Xinjiang, saying the information is “commercially sensitive.” But last year, the US denim mill Cone Denim announced that it used Oritain’s technology for an audit it said showed, “with scientific certainty, that all samples analyzed across Cone Denim operations in China and Mexico were compliant and zero samples were identified as coming from a restricted risk origin as stated in Cone Denim’s cotton sourcing policy, which includes the Xinjiang region of China, Syria, Uzbekistan and Turkmenistan.”

Cone Denim declined to be interviewed for this story.

While Cochrane said he sees Oritain’s technology as scalable with the possibility for widespread adoption, he pointed out forensic science isn’t a simple process. It requires more than one-off testing to audit a supply chain.

“We look at ourselves as one tool in the toolkit to ensure integrity,” he said.

It’s also the case that algorithmic models always have some margin of error.

“We can say it’s really accurate if we’ve seen that exact thing before,” said Tim Wilson, whose company, 3P Strategy, advises clients on supply-chain transparency. “But as soon as it’s not something we’ve specifically seen before, then the accuracy goes adrift.”

No Shortcuts Doesn’t Mean No Value

Wilson pointed to other complications. A DNA test may not satisfy regulators looking to document the full history of a garment, as cotton might be grown in one country then shipped to China where it might be spun into yarn or milled into fabric in Xinjiang.

Testing still has value, however. Methods that allow anyone to easily and accurately determine the origin of materials in a finished product “help a lot with traceability and verifying the other due diligence work that we’re doing to get back to the raw material in our supply chains,” AAFA’s Herman said.

And the work is ongoing to make these technologies more effective. In Wan’s view, if the cotton industry dedicated the resources to the project, it could build an open-source database of cotton samples from around the world in less than two years.

That’s probably a long shot given the cooperation required when companies and organizations in the industry may have competing interests. But the pressure on brands and retailers to be able to trace raw materials back to their source isn’t set to decrease anytime soon. It will likely keep demand and investment moving toward technologies that can solve even part of the problem.