A recent study published in the Proceedings of the National Academy of Sciences highlights the innovative use of artificial intelligence to identify potential antimicrobial compounds for tuberculosis treatment. The research was led by scientists from the University of California San Diego, Linnaeus Bioscience Inc., and the Center for Global Infectious Disease Research at the Seattle Children’s Research Institute.
A fluorescence microscopy image reveals the tuberculosis-causing bacterium Mycobacterium tuberculosis after an antimicrobial treatment. Membranes are stained red, DNA blue and areas of membrane permeability appear green. These dramatic changes in bacterial cell structure form consistent patterns that can be used to identify how potential new treatments work — a critical step in developing effective therapies for the globally significant tuberculosis disease. Image Credit: UC San Diego
A major threat to world health, tuberculosis infected over 10 million people in 2022. The pathogen responsible for "TB" can cause persistent coughing, chest pains, exhaustion, fever, and weight loss when it spreads through the air and into the lungs.
Other regions of the world have more widespread infections, but Kansas is currently experiencing one of the worst TB outbreaks in US history, which has sparked two fatalities.
Antibiotics are normally used to treat tuberculosis, but the emergence of drug-resistant strains has made the need for novel treatment candidates urgent.
Linnaeus Bioscience, a biotechnology company based in San Diego, was founded on technology developed in the UC San Diego School of Biological Sciences by Professor Joe Pogliano and Dean Kit Pogliano. Their bacterial cytological profiling (BCP) method offers a rapid approach to understanding antibiotic function by quickly identifying their mechanisms of action.
The search for new tuberculosis drug targets using traditional laboratory methods has been challenging and time-consuming, largely due to the complexity of understanding how new drugs act against Mycobacterium tuberculosis, the bacterium responsible for the disease.
The development of "MycoBCP," a next-generation technology funded by the Gates Foundation, is detailed in the recent PNAS study. To overcome conventional obstacles and provide fresh perspectives on Mycobacterium tuberculosis cells, the novel approach combines BCP with deep learning, a form of artificial intelligence that makes use of neural networks that resemble brains.
This is the first time that this kind of image analysis using machine learning and AI has been applied in this way to bacteria. Tuberculosis images are inherently difficult to interpret by the human eye and traditional lab measurements. Machine learning is much more sensitive in being able to pick up the differences in shapes and patterns that are important for revealing underlying mechanisms.
Joe Pogliano, Professor and Study Co-Author, Department of Molecular Biology, University of California, San Diego
The MycoBCP technology was developed over two years by study lead authors Joseph Sugie and Diana Quach, who trained AI tools called convolutional neural networks using over 46,000 images of TB cells. Both Sugie and Quach are currently at Linnaeus Bioscience, and they both received PhDs from the Shu Chien-Gene Lay Department of Bioengineering and finished postdoctoral appointments in the Department of Molecular Biology's Pogliano labs.
Tuberculosis cells are clumpy and seem to always stick close to each other, so defining cell boundaries didn’t seem possible. Instead, we jumped straight into letting the computer analyze the patterns in the images for us.
Joseph Sugie, Chief Technology Officer and Study Lead Author, Linnaeus Bioscience
Linnaeus collaborated with Tanya Parish, a tuberculosis specialist at Seattle Children's Research Institute, to create BCP for mycobacteria. In addition to helping the team find the best candidate compounds for drug development, the new system has already significantly increased the team's capacity for TB research.
A critical component of progressing towards new drug candidates is defining how they work, which has been technically challenging and takes time. This technology expands and accelerates our ability to do this and allows us to prioritize which molecules to work on based on their mode of action. We were excited to collaborate with Linnaeus in their work to develop this technology to M. tuberculosis.
Tanya Parish, Study Co-Author and Tuberculosis Specialist, Seattle Children's Research Institute
UC San Diego Biotech Spinoff Tackles Global Health Issue
With a technology created at UC San Diego, Linnaeus Bioscience was founded in 2012 with the goal of revolutionizing one’s knowledge of how antibiotics function.
“We developed bacterial cytological profiling and it allowed us to look at bacterial cells in a new way. It allowed us to really see how cells look after treatment with antibiotics so we could interpret their underlying mechanisms. We describe this method as equivalent to performing an autopsy on a bacterial cell,” said Joe Pogliano.
Joe and Kit Pogliano introduced the BCP technology into the market by establishing Linnaeus Bioscience in San Diego's local biotechnology hub. This gave other businesses access to the technology. Samples from all over the world are now sent to the company for quick analysis and identification of novel bacterial drug candidates.
Pogliano attributes the company's success and expansion to the biotechnology community, particularly to its early residence in the San Diego JLABS incubator, which helps start-up biotech businesses.
“We could not have gotten Linnaeus Bioscience off the ground if not for the supportive biotech community and the infrastructure provided at JLABS. All of the company’s employees at Linnaeus obtained their Ph.Ds at UC San Diego so this has become a great UC San Diego research, alumni and San Diego biotech community success story, culminating in this new AI platform to help solve the antibiotic resistance crisis,” said Pogliano.
The study coauthors include Marc Sharp, Sara Ahmed, Lauren Ames, Amala Bhagwat, Aditi Deshpande, and Tanya Parish in addition to Quach, Pogliano, and Sugie.
Funding for the study came from the Bill and Melinda Gates Foundation.
Journal Reference:
Quach, D., et al. (2025) Deep learning–driven bacterial cytological profiling to determine antimicrobial mechanisms in Mycobacterium tuberculosis. Proceedings of the National Academy of Sciences. doi.org/10.1073/pnas.2419813122