Communications of the ACM
Software Helps Spot New Forms of Fentanyl and Other Illegal Drugs
Credit: 3DChem.com
Fentanyl, the synthetic drug that is driving a nationwide overdose epidemic, is not only a killer, it's also a shape shifter. Illicit chemists are constantly cooking up new forms of fentanyl, each with a slightly different chemical structure, stymieing law enforcement and putting users at greater risk.
To control fentanyl, which mimics heroin but is more potent, forensic chemists need to identify it. But when they encounter a new type of fentanyl, called a fentanyl analog, it will not yet be in the chemical databases they use to identify illegal drugs. Now, the National Institute of Standards and Technology (NIST), an agency of the U.S. Department of Commerce, has released a free software tool to help.
The NIST tool contains an algorithm for searching chemical databases that can recognize new fentanyl analogs even if there are no matches in the database. This method, called Hybrid Similarity Search, works with a common laboratory technique called mass spectrometry and is described in "Combining Fragment-Ion and Neutral-Loss Matching during Mass Spectral Library Searching: A New General Purpose Algorithm Applicable to Illicit Drug Identification," published in the journal Analytical Chemistry.
"If you search for one compound, you will find all the compounds that have a similar chemical structure," says Arun Moorthy, a NIST postdoc fellow and mathematical statistician who worked on the algorithm, and an author of the Analytical Chemistry paper. "That should help law enforcement and public health authorities react more quickly when a new and deadly drug hits the streets."
The method also works with synthetic cathinones—more commonly known as "bath salts"—synthetic marijuana, and other drugs.
Suspect's Fingerprints
When crime lab chemists receive a bag of powder that might contain illegal drugs, their first step, like any criminal investigator, is to fingerprint the suspect. In this case, they get "molecular fingerprints" of whatever is in the powder, then run those fingerprints against a database of known suspects to look for a likely match.
To get those fingerprints, they insert a sample of the powder into an instrument called a mass spectrometer, which bombards the sample with electrons. This shatters the molecules into fragments. Then it sorts those fragments by their weight, or mass, to produce a "mass spectrum"—a unique pattern of vertical lines that can be used as a molecular fingerprint.
One of the most commonly used databases of molecular fingerprints is the Mass Spectral Library maintained by the Scientific Working Group for the Analysis of Seized Drugs, or SWGDRUG. If forensic chemists search that database, which currently contains 88 fentanyl analogs, and get a hit, they will do further tests to confirm the identification. But if they don't get a hit, the substance will remain unknown.
Whack-a-Mole
Why do illicit chemists invent new fentanyl analogs? One reason is that tweaking the structure of the compound can enhance its narcotic effect, producing what users might consider a better high. Another reason is to dodge law enforcement. Before presenting evidence about a new analog in court, forensic chemists need to discover it and work out its chemical structure.
"Putting a new molecule in the books takes time, and before you know it, there's another one out there," says Sandra Rodriguez-Cruz, a senior research chemist with the Drug Enforcement Administration's Southwest Laboratory. "It's a Whack-a-Mole game."
To create a new analog, illicit chemists change some of the atoms in the molecule while leaving the core structure intact. They might add a chlorine atom to one branch or remove a hydrogen from another. This almost always changes the compound's fingerprint by shifting some of the lines in the mass spectrum.
"Our algorithm corrects for those shifts, so you can find related compounds," says Stephen Stein, the NIST research chemist who oversaw the development of the algorithm, and an author of the published paper.
An experienced chemist can also correct for those changes manually. But the manual method takes time and has to be done separately for each known compound that the unknown might be related to.
"The concept is intuitive to chemists, but it has never been captured in an algorithm before," Moorthy says. Now that it has, a computer can churn through an entire database looking for related compounds.
The Hybrid Similarity Search has been built into the NIST 17 MS Search software, which can be downloaded from a NIST webpage. Users can submit the mass spectrum for an unknown compound. The program will return a list of the most closely related compounds. If that list contains fentanyl or any of its analogs, the unknown might be a fentanyl analog as well. That list can provide a head start on elucidating the new compound's chemical structure.
"Hybrid search is not the silver bullet that will solve the opioid epidemic, but it is a very useful tool," Rodriguez-Cruz says. "If you have a difficult molecule, it can speed up your workflow significantly."
Additional authors of the Analytical Chemistry paper are NIST's William E. Wallace, Anthony J. Kearsley, and Dmitrii V. Tchekhovskoi.
No entries found