How can machine learning make a difference in healthcare and the environment? Researchers from IBM and UC San Francisco will use the technology's algorithms to uncover how to modify cells to detect and treat diseases, and develop biofuels and environmentally responsive plants.
On Monday, the National Science Foundation awarded UC San Francisco $24 million over five years for the creation of an interdisciplinary bioengineering center. At the new Center for Cellular Construction, researchers in biology, engineering, physics, and computer science from San Francisco State University, Stanford University, and UC Berkeley will use cellular engineering to design cells that act as machines.
Scientists will use IBM Watson's AI platform to study the connection between how cells structure themselves and the functions they perform. Ultimately, the goal is to create tools that "allow scientists to reliably engineer cells for desired functions by modifying their internal organization," according to a press release.
"By uncovering the rules of a cell's organization, we think we can understand what the cell is sensing, and use that for a number of situations," said Simone Bianco, research staff member in the department of industrial and applied genomics at the IBM Almaden Research Center.
According to the press release, the Center for Cellular Construction's five projects will integrate basic research and real-world applications:
- The Cellular Machine Shop, a core facility with cell engineering tools including imaging systems and gene sequencing machines
- Computer Aided Design, which will create a computational platform that can design cells
- Multicellular Engineering, which will build molecular tools that can construct multicellular structures for more complex functions—for example, designing algae that can store more energy and defend itself from other microbes.
- Living Bioreactor, in which scientists will modify cells' internal structures and combine them with other organisms and cell types to change cell behavior
- Cell State Interference Engine"We can use this to monitor how the cell looks in real time, to detect early signs of anomalous conditions." Simone Bianco, research staff member in the department of industrial and applied genomics at the IBM Almaden Research Center, which Bianco will lead, designing image analysis software that will use engineered cells as living sensors of environmental conditions, such as air pollution.
Bianco's team is using IBM Watson's machine learning algorithms to scan hundreds of thousands of images of healthy cells to establish a baseline. Then, they will examine what the cell looks like when they introduce a specific outside entity, such as a cancer-causing cell or an invasive species.
"We can use this to monitor how the cell looks in real time, to detect early signs of anomalous conditions," Bianco said.
Currently, if a doctor finds a potential tumor, a patient must undergo a biopsy. The pathologist looks at the image of the tissue and compares it to images of normal cells to make a diagnosis. "When we get to the level of the pathologist, there is already something messed up in the cells," Bianco said. "We're looking for early modifications of the morphology of the cell that will help the doctor make an early decision or a better decision." Someday, this kind of cell screening could be done in a routine test, Bianco said.
If the team can understand how cells regulate their behavior, it can eventually try to engineer cells for different purposes, Bianco said. "You can imagine engineering cells that react in specific ways to the presence of cancer," he said. "This will be useful for diagnostics, but also in principle for therapeutics."
But modifying cells for use in environmental sensing will likely be common years before doing so to treat cancer, Bianco said. For example, his team will study how microorganisms in bodies of water react to the presence of a pollutant or invasive species. This can help better engineer them to either remove pollution or defend themselves, among a variety of other applications.
"The most important thing is to try and understand cellular behavior," Bianco said. "If we have the full command of that, there are virtually no limits to the potential technology that can come out of this basic research. It can be applied anywhere."
For example, you might be able to bring bacteria to space and see how it reacts to the atmosphere of other planets, he said.
The Center for Cellular Construction represents a new paradigm in technology research, Bianco said. "What we see more and more, especially in Silicon Valley, is that disciplines that were separated are now becoming integrated," he said. "Physicists are working with biologists and engineers and computer scientists. The partnerships within academia and industries are really the future of research."
The 3 big takeaways for TechRepublic readers
- On Monday, the National Science Foundation awarded UC San Francisco $24 million to build the Center for Cellular Construction, which will bring together researchers from IBM and other academic institutions to study cellular engineering and its real-world applications.
- One of the center's five projects will use IBM Watson's machine learning algorithms to study how healthy cells react when certain environmental conditions are introduced, to eventually engineer cells for different purposes.
- Cellular engineering could eventually be used to protect the environment, and to detect and treat diseases such as cancer at the earliest stages.
- New IBM lab-on-a-chip uses nanotechnology to detect cancer, Zika virus faster (TechRepublic)
- IBM's Watson enters drone business courtesy of deal with Aerialtronics (ZDNet)
- IBM ups enterprise appeal of Swift with new Bluemix runtime (TechRepublic)
- IBM gets R big data on-ramp for Watson via Columbus Collaboratory (ZDNet)
- MIT's answer to global health issues: Democratizing big data analytics (TechRepublic)
Alison DeNisco is a Staff Writer for TechRepublic. She covers CXO and the convergence of tech and the workplace.