The research was published in Cell Metabolism and reported by The Scientist, a sibling brand of Dark Daily.
For clinical laboratory professionals, the approach highlights a possible shift away from time-intensive stool-based sequencing toward faster, more scalable testing workflows. “One of the key barriers to integrating our knowledge of the microbiome into clinical care is the time it takes to analyze the data on the microbiome,” said Ariel Hernandez-Leyva, an MD/PhD student working for gut microbiome researcher Andrew Kau’s group at Washington University School of Medicine. (Photo credit: Kau Lab)
Breath-Based Testing Could Streamline Microbiome Workflows
The research team found that volatile organic compounds (VOCs) in breath samples closely correlate with gut microbiome activity, suggesting a streamlined alternative that could reduce turnaround times and expand access to microbiome-informed diagnostics. In both human and mouse studies, breath “volatilome” profiles mirrored microbial metabolites in the gut.
In a proof-of-concept analysis, VOC patterns distinguished children with asthma from healthy controls and predicted levels of a gut bacterial species associated with the condition. Such capabilities could support earlier clinical decision-making and reduce reliance on complex sequencing workflows.
If validated in larger studies, breath-based diagnostics could offer clinical labs a practical pathway to integrate microbiome insights into routine testing, with potential applications in pediatric care, infectious disease risk assessment, and chronic disease management.
This article was created with the assistance of Generative AI and has undergone editorial review before publishing.
AI-designed molecular sensors could enable ultra-early cancer detection through simple urine tests, signaling major shifts ahead for clinical laboratories and diagnostic workflows.
Artificial intelligence (AI) is beginning to reshape how cancer could be detected and that shift may carry significant implications for clinical laboratories. Researchers at MIT and Microsoft have developed an AI-driven system that designs molecular sensors capable of detecting cancer-linked enzyme activity at extremely early stages, potentially through a simple urine test that could one day be used at home.
The approach centers on proteases, enzymes that are often overactive in cancer and play a role in tumor growth and metastasis. For more than a decade, researchers have explored the idea of using protease activity as a biomarker. Now, AI is accelerating that work by improving the precision and scalability of sensor design.
“We’re focused on ultra-sensitive detection in diseases like the early stages of cancer, when the tumor burden is small, or early on in recurrence after surgery,” said Sangeeta Bhatia, professor of health sciences and technology at MIT and senior author of the study, published in Nature Communications.
From Trial-and-Error Peptides to AI-Optimized Protease Sensors
The researchers coat nanoparticles with short protein sequences, or peptides, that are engineered to be cleaved by specific proteases. When these nanoparticles travel through the body and encounter cancer-associated proteases, the peptides are cut and excreted in urine, where the signal can be detected using a simple paper strip. The pattern of signals could indicate not only the presence of cancer but also its type.
Earlier versions of this technology relied on trial-and-error methods to identify peptides,
often resulting in signals that were not specific to a single protease. While multiplexed peptide panels still produced diagnostic signatures in animal models, they lacked enzyme-level specificity—an important limitation for clinical translation.
The new AI system, called CleaveNet, is designed to overcome that challenge. Using a protein “language model,” CleaveNet can generate peptide sequences optimized for both efficiency and specificity against a target protease.
“If we know that a particular protease is really key to a certain cancer, and we can optimize the sensor to be highly sensitive and specific to that protease, then that gives us a great diagnostic signal,” said Ava Amini, a principal researcher at Microsoft Research. (Photo credit: Microsoft)
For lab leaders, the implications are significant. AI-designed sensors could reduce assay complexity, improve signal clarity, and lower development costs by narrowing the number of biomarkers needed for reliable detection. They also hint at a future where decentralized, at-home testing complements centralized laboratory diagnostics, shifting labs toward validation, data interpretation, and longitudinal disease monitoring.
Bhatia’s lab is now part of an Advanced Research Projects Agency for Health–funded effort to develop an at-home diagnostic capable of detecting up to 30 cancer types in early stages. Beyond diagnostics, the same AI-designed peptides could be incorporated into targeted therapeutics, releasing drugs only within tumor environments.
As AI-driven biomarker discovery advances, clinical laboratories may find themselves at the center of integrating these technologies into regulated testing pathways—reshaping early cancer detection and redefining the lab’s role in precision oncology.
A new partnership between Intermountain Health and Testmate Health aims to bring rapid, lab-quality STI testing out of the central lab and into underserved communities, addressing persistent gaps in diagnosis and follow-up care.
Testmate Health and Intermountain Health have entered a strategic partnership and investment aimed at accelerating access to rapid, low-cost molecular testing for sexually transmitted infections (STIs) across the US.
For clinical laboratory leaders, the collaboration signals a growing push to move high-quality molecular diagnostics closer to patients, particularly those belonging to underserved and high-risk populations.
STIs continue to represent a major public health challenge, with an estimated 80% of chlamydia and gonorrhea infections going undiagnosed each year. Delays in testing and treatment are especially common among college students, LGBTQ+ populations, and patients served by rural or resource-limited clinics. The two organizations say their partnership is designed to close those gaps by making accurate, lab-quality testing available outside of traditional laboratory environments.
Bringing Molecular Diagnostics Beyond the Central Lab
Under the agreement, Intermountain Health will support the deployment of Testmate’s single-use, reader-free molecular tests for Chlamydia trachomatis and Neisseria gonorrhoeae. The tests are designed to deliver results in under 30 minutes and do not require central lab infrastructure. They can be used with urine or swab samples, offering flexibility for a range of care settings.
Karen Brownell, vice president of Lab Services at Intermountain Health said, “When these STI tests become FDA-approved in the US, Testmate’s innovative approach to molecular diagnostics will allow us to deliver lab-quality results outside traditional lab settings, directly impacting communities that have historically lacked access to timely testing.” (Photo credit: ContactOut)
Addressing Access, Turnaround Time, and Follow-Up
Testmate’s leadership emphasized that reducing barriers to testing is central to improving outcomes. Rapid turnaround times may help clinicians initiate treatment during the same visit, reducing loss to follow-up—a persistent issue in STI management.
The organizations say combining Testmate’s physician-developed diagnostics with Intermountain’s clinical infrastructure could also lower overall healthcare costs by enabling earlier detection and treatment.
For laboratory leaders, the collaboration highlights a broader trend toward decentralized molecular testing and point-of-care strategies that complement, rather than replace, core laboratory services. As health systems look to improve access and equity while managing costs, partnerships like this one may foreshadow how labs extend their impact beyond traditional walls.
