News, Analysis, Trends, Management Innovations for
Clinical Laboratories and Pathology Groups

Hosted by Robert Michel

News, Analysis, Trends, Management Innovations for
Clinical Laboratories and Pathology Groups

Hosted by Robert Michel
Sign In

McMaster University Uses AI Algorithm, Machine Learning to Find Antibiotic That Neutralizes Common Antimicrobial Resistant Superbug

Further development of this novel technology could result in new, more sensitive assays for clinical laboratories to use in the effort to improve antimicrobial stewardship in hospitals

Researchers at McMaster University in Ontario, Canada, have used artificial intelligence (AI) to identify a potential antibiotic that neutralizes the drug-resistant bacteria Acinetobacter baumannii, an antibiotic resistant pathogen commonly found in many hospitals. This will be of interest to clinical laboratory managers and microbiologists involved in identifying strains of bacteria to determine if they are antimicrobial-resistant (AMR) superbugs.

Using machine learning, the scientists screened thousands molecules to look for those that inhibited the growth of this specific pathogen. And they succeeded.

“We trained a neural network with this growth inhibition dataset and performed in silico predictions for structurally new molecules with activity against A. baumannii,” the researchers wrote in their published study.

They discovered that the molecule abaucin inhibited the growth of the antibiotic-resistant pathogen in vitro.

This shows how machine learning and AI technologies are giving biomedical researchers  tools to identify new therapeutic drugs that are effective against drug-resistant strains of bacteria. This same research can be expected to lead to new clinical laboratory assays that determine if superbugs can be attacked by specific therapeutic drugs.

The researchers published their findings in the journal Nature Chemical Biology titled, “Deep Learning-Guided Discovery of an Antibiotic Targeting Acinetobacter Baumannii.”

“When I think about AI in general, I think of these models as things that are just going to help us do the thing we’re going to do better,” Jonathan Stokes, PhD, Assistant Professor of Biomedicine and Biochemistry at McMaster University in Ontario, Canada, and lead author of the study, told USA Today. Clinical laboratory scientists and microbiologists will be encouraged by the McMaster University scientists’ findings. (Photo copyright: McMaster University.)

McMaster Study Details

Jonathan Stokes, PhD, head of the Stokes Laboratory at McMaster University, is Assistant Professor of Biomedicine/Biochemistry at McMaster and lead author of the study. Stokes’ team worked with researchers from the Broad Institute of MIT and Harvard to explore the effectiveness of AI in combating superbugs, USA Today reported.

“This work highlights the utility of machine learning in antibiotic discovery and describes a promising lead with targeted activity against a challenging Gram-negative pathogen,” the researchers wrote in Nature Chemical Biology.

Stokes Lab utilized the high-throughput drug screening technique, spending weeks growing and exposing Acinetobacter baumannii to more than 7,500 agents of drugs and active ingredients of drugs. When 480 compounds were uncovered that blocked the growth of bacteria, this information was then provided to a computer that was trained to run an AI algorithm, CNN reported.

“Once we had our [machine learning] model trained, what we could do then is start showing that model brand-new pictures of chemicals that it had never seen, right? And based on what it had learned during training, it would predict for us whether those molecules were antibacterial or not,” Stokes told CNN.

The model spent hours screening more than 6,000 molecules. It then narrowed the search to 240 chemicals, which were tested in the lab. The scientists pared down the results to the nine most effective inhibitors of bacteria. They then eliminated those that were either related to existing antibiotics or might be considered dangerous.

The researchers found one compound—RS102895 (abaucin)—which, according to Stokes, was likely created to treat diabetes, CNN reported. The scientists discovered that the compound prevented bacterial components from making their way from inside a cell to the cell’s surface.

“It’s a rather interesting mechanism and one that is not observed amongst clinical antibiotics so far as I know,” Stokes told CNN.

Because of the effectiveness of the antibiotic during testing on mice skin, the researchers believe this method may be useful for creating antibiotics custom made to battle additional drug resistant pathogens, CNN noted.

Defeating a ‘Professional Pathogen’

Acinetobacter baumannii (A. baumannii)—the focus of Stoke’s study—is often found on hospital counters and doorknobs and has a sneaky way of using other organisms’ DNA to resist antibiotic treatment, according to CNN

“It’s what we call in the laboratory a professional pathogen,” Stokes told CNN.

A. baumannii causes infections in the urinary tract, lungs, and blood and typically wreaks havoc to vulnerable patients on breathing machines, in intensive care units, or undergoing surgery, USA Today reported.

A. baumannii is resistant to carbapenem, a potent antibiotic. The Centers for Disease Control and Prevention (CDC) reported that in 2017 the bacteria infected 8,500 people in hospitals, 700 of those infections being fatal.

Further, in its 2019 “Antibiotic Resistance Threats in the United States” report, the CDC stated that one out of every four patients infected with the bacteria died within one month of their diagnosis. The federal agency deemed the bacteria “of greatest need” for new antibiotics.

Thus, finding a way to defeat this particularly nasty bacteria could save many lives.

Implications of Study Findings on Development of new Antibiotics

The Stokes Laboratory study findings show promise. If more antibiotics worked so precisely, it’s possible bacteria would not have a chance to become resistant in the first place, CNN reported.

Next steps in Stokes’ research include optimizing the chemical structure and testing in larger animals or humans, USA Today reported.

“It’s important to remember [that] when we’re trying to develop a drug, it doesn’t just have to kill the bacterium,” Stokes noted. “It also has to be well tolerated in humans and it has to get to the infection site and stay at the infection site long enough to elicit an effect,” USA Today reported.

Stokes’ study is a prime example of how AI can make a big impact in clinical laboratory diagnostics and treatment.

“We know broad-spectrum antibiotics are suboptimal and that pathogens have the ability to evolve and adjust to every trick we throw at them … AI methods afford us the opportunity to vastly increase the rate at which we discover new antibiotics, and we can do it at a reduced cost. This is an important avenue of exploration for new antibiotic drugs,” Stokes told CNN.

Clinical laboratory managers and microbiologists may want to keep an open-mind about the use of AI in drug development. More research is needed to give substance to the McMaster University study’s findings. But the positive results may lead to methods for fine tuning existing antibiotics to better combat antimicrobial-resistant bacteria, USA Today reported.

—Kristin Althea O’Connor

Related Information:

The Study: Deep Learning-Guided Discovery of an Antibiotic Targeting Acinetobacter Baumannii

Scientists Use AI to Discover Antibiotic to Fight Deadly Hospital Bug

A New Type of Antibiotic, Discovered with Artificial Intelligence, May Defeat a Dangerous Superbug

WHO Report: Bacteria for Which New Antibiotics are Urgently Needed

Abstract on Deaths of Those Infected with Acinetobacter

CDC: Antibiotic Resistance Threats in the United States

CDC: Acinetobacter in Healthcare Settings

Home Ice: Star Researcher Back at Mac to Pioneer Use of AI to Fight Antibiotic Resistance

Diagnosing Ovarian Cancer Using Perception-based Nanosensors and Machine Learning

Two studies show the accuracy of perception-based systems in detecting disease biomarkers without needing molecular recognition elements, such as antibodies

Researchers from multiple academic and research institutions have collaborated to develop a non-conventional machine learning-based technology for identifying and measuring biomarkers to detect ovarian cancer without the need for molecular identification elements, such as antibodies.

Traditional clinical laboratory methods for detecting biomarkers of specific diseases require a “molecular recognition molecule,” such as an antibody, to match with each disease’s biomarker. However, according to a Lehigh University news release, for ovarian cancer “there’s not a single biomarker—or analyte—that indicates the presence of cancer.

“When multiple analytes need to be measured in a given sample, which can increase the accuracy of a test, more antibodies are required, which increases the cost of the test and the turnaround time,” the news release noted.

The multi-institutional team included scientists from Memorial Sloan Kettering Cancer Center, Weill Cornell Medicine, the University of Maryland, the National Institutes of Standards and Technology, and Lehigh University.

Unveiled in two sequential studies, the new method for detecting ovarian cancer uses machine learning to examine spectral signatures of carbon nanotubes to detect and recognize the disease biomarkers in a very non-conventional fashion.

Daniel Heller, PhD
 
“Carbon nanotubes have interesting electronic properties,” said Daniel Heller, PhD (above), in the Lehigh University news release. “If you shoot light at them, they emit a different color of light, and that light’s color and intensity can change based on what’s sticking to the nanotube. We were able to harness the complexity of so many potential binding interactions by using a range of nanotubes with various wrappings. And that gave us a range of different sensors that could all detect slightly different things, and it turned out they responded differently to different proteins.” This method differs greatly from traditional clinical laboratory methods for identifying disease biomarkers. (Photo copyright: Memorial Sloan-Kettering Cancer Center.)

Perception-based Nanosensor Array for Detecting Disease

The researchers published their findings from the two studies in the journals Science Advances, titled, “A Perception-based Nanosensor Platform to Detect Cancer Biomarkers,” and Nature Biomedical Engineering, titled, “Detection of Ovarian Cancer via the Spectral Fingerprinting of Quantum-Defect-Modified Carbon Nanotubes in Serum by Machine Learning.”

In the Science Advances paper, the researchers described their development of “a perception-based platform based on an optical nanosensor array that leverages machine learning algorithms to detect multiple protein biomarkers in biofluids.

“Perception-based machine learning (ML) platforms, modeled after the complex olfactory system, can isolate individual signals through an array of relatively nonspecific receptors. Each receptor captures certain features, and the overall ensemble response is analyzed by the neural network in our brain, resulting in perception,” the researchers wrote.

“This work demonstrates the potential of perception-based systems for the development of multiplexed sensors of disease biomarkers without the need for specific molecular recognition elements,” the researchers concluded.

In the Nature Biomedical Engineering paper, the researchers described a fined-tuned toolset that could accurately differentiate ovarian cancer biomarkers from biomarkers in individuals who are cancer-free.

“Here we show that a ‘disease fingerprint’—acquired via machine learning from the spectra of near-infrared fluorescence emissions of an array of carbon nanotubes functionalized with quantum defects—detects high-grade serous ovarian carcinoma in serum samples from symptomatic individuals with 87% sensitivity at 98% specificity (compared with 84% sensitivity at 98% specificity for the current best [clinical laboratory] screening test, which uses measurements of cancer antigen 125 and transvaginal ultrasonography,” the researchers wrote.

“We demonstrated that a perception-based nanosensor platform could detect ovarian cancer biomarkers using machine learning,” said Yoona Yang, PhD, a postdoctoral research associate in Lehigh’s Department of Chemical and Biomolecular Engineering and co-first author of the Science Advances article, in the news release.

How Perception-based Machine Learning Platforms Work

According to Yang, perception-based sensing functions like the human brain.

“The system consists of a sensing array that captures a certain feature of the analytes in a specific way, and then the ensemble response from the array is analyzed by the computational perceptive model. It can detect various analytes at once, which makes it much more efficient,” Yang said.

The “array” the researchers are referring to are DNA strands wrapped around single-wall carbon nanotubes (DNA-SWCNTs).

“SWCNTs have unique optical properties and sensitivity that make them valuable as sensor materials. SWCNTS emit near-infrared photoluminescence with distinct narrow emission bands that are exquisitely sensitive to the local environment,” the researchers wrote in Science Advances.

“Carbon nanotubes have interesting electronic properties,” said Daniel Heller, PhD, Head of the Cancer Nanotechnology Laboratory at Memorial Sloan Kettering Cancer Center and Associate Professor in the Department of Pharmacology at Weill Cornell Medicine of Cornell University, in the Lehigh University news release.

“If you shoot light at them, they emit a different color of light, and that light’s color and intensity can change based on what’s sticking to the nanotube. We were able to harness the complexity of so many potential binding interactions by using a range of nanotubes with various wrappings. And that gave us a range of different sensors that could all detect slightly different things, and it turned out they responded differently to different proteins,” he added.

The researchers put their technology to practical test in the second study. The wanted to learn if it could differentiate symptomatic patients with high-grade ovarian cancer from cancer-free individuals. 

The research team used 269 serum samples. This time, nanotubes were bound with a specific molecule providing “an extra signal in terms of data and richer data from every nanotube-DNA combination,” said Anand Jagota PhD, Professor, Bioengineering and Chemical and Biomolecular Engineering, Lehigh University, in the news release.

This year, 19,880 women will be diagnosed with ovarian cancer and 12,810 will die from the disease, according to American Cancer Society data. While more research and clinical trials are needed, the above studies are compelling and suggest the possibility that one day clinical laboratories may detect ovarian cancer faster and more accurately than with current methods.   

—Donna Marie Pocius

Related Information:

Perception-Based Nanosensor Platform Could Advance Detection of Ovarian Cancer

Perception-Based Nanosensor Platform to Detect Cancer Biomarkers

Detection of Ovarian Cancer via the Spectral Fingerprinting of Quantum-Defect-Modified Carbon Nanotubes in Serum by Machine Learning

Machine Learning Nanosensor Platform Detects Early Cancer Biomarkers

;