Researchers from Carnegie Mellon University's Computational Biology Department in the School of Computer Science have developed a new process that could reinvigorate the search for natural product drugs to treat cancers, viral infections, and other ailments.

The machine learning algorithms developed by the Metabolomics and Metagenomics Lab match the signals of a microbe's metabolites with its genomic signals and identify which likely correspond to a natural product. Knowing that researchers are better equipped to isolate the natural product to begin developing it for a possible drug.

"Natural products are still one of the most successful paths for drug discovery," said Bahar Behsaz, a project scientist in the lab and lead author of a paper about the process. "And we think we're able to take it further with an algorithm like ours. Our computational model is orders of magnitude faster and more sensitive." 

In a single study, the team was able to scan the metabolomics and genomic data for about 200 strains of microbes. The algorithm not only identified the hundreds of natural product drugs the researchers expected to find, but it also discovered four novel natural products that appear promising for future drug development. 

The team has developed NRPminer, an artificial intelligence tool to aid in discovering non-ribosomal peptides (NRPs). NRPs are an important type of natural product and are used to make many antibiotics, anticancer drugs, and other clinically used medications. They are, however, difficult to detect and even more difficult to identify as potentially useful.

"What is unique about our approach is that our technology is very sensitive. It can detect molecules with nanograms of abundance," said Hosein Mohimani, an assistant professor and head of the lab. "We can discover things that are hidden under the grass."

Most of the antibiotic, antifungal, and many antitumor medications discovered and widely used have come from natural products.

Penicillin is among the most used and well-known drugs derived from natural products. It was, in part, discovered by luck, as are many of the drugs made from natural products. But replicating that luck is difficult in the laboratory and at scale. Trying to uncover natural products is also time and labor-intensive, often taking years and millions of dollars. Major pharmaceutical companies have mostly abandoned the search for new natural products in the past decades.

By applying machine learning algorithms to the study of genomics, however, researchers have created new opportunities to identify and isolate natural products that could be beneficial.

"Our hope is that we can push this forward and discover other natural drug candidates and then develop those into a phase that would be attractive to pharmaceutical companies," Mohimani said. "Bahar Behsaz and I are expanding our discovery methods to different classes of natural products at a scale suitable for commercialization."

The team is already investigating the four new natural products discovered during their study. The products are being analyzed by a team led by Helga Bode, head of the Institute for Molecular Bioscience at Goethe University in Germany, and two have been found to have potential antimalarial properties.

If you take a bucket of water balloons and jostle one of them, the neighboring balloons will respond as well. This is a scaled-up example of how collections of cells and other deformable particle packings respond to forces. Modeling this phenomenon with supercomputer simulations can shed light on questions about how cancer cells invade healthy tissue or how leaves and flowers grow. But the behavior of cell aggregates is extremely complex, and fully capturing their structure and dynamics has proved tricky. 

A team of researchers in the lab of Corey O’Hern, professor of mechanical engineering & materials science, physics, and applied physics, has developed novel supercomputer simulations of deformable particles that more accurately model their collective behavior. The study was led by John Treado, a Ph.D. student, and postdoctoral researcher Dong Wang, both in the O’Hern lab. It was recently published in Physical Review Materials.  

Cells, bubbles, droplets, and other small particles that make up soft solids – which include anything from mayonnaise and shaving cream to cells and tissues - are all highly deformable. There’s significant variability in how they change shape, and how they respond to forces.  

“There is a strong connection between the response of the collection of particles to applied forces, particle shape, and deformability,” Treado said. “Particle deformability determines how they’re going to move because they’re compressed tightly with many neighbors who are squishing them on all sides.” 

Conventional computer models typically represent soft particles as spheres. When the spheres press against each other, the models represent the spheres’ deformations by having them overlap. This approach works to a certain extent, but crucial information about the particle shapes and interactions is lost or misrepresented. 

The O’Hern team, though, developed a supercomputer model that can tune the particles from being floppy, with the ability to easily change shape, to being completely rigid. This model treats each particle as a ring of connected small spheres. In the simulation, forces are applied to the spherical beads, and the model tracks how the connected beads change positions and orientations. 

The researchers found that allowing for collective shape changes produced material responses that they wouldn’t have observed with fixed spherical shapes of the particles. The results underscore the importance of incorporating shape variability into models of tissues, foams, and other soft solids composed of deformable particles.

“We now need to extend the model to three dimensions, which more closely mimics the real world,” Wang said. “We can also apply the deformable particle model to active biological systems, which can form swarms, schools, and flocks.” 

Treado and Wang are also currently using this new supercomputer model to study how tumor cells invade adipose tissue in breast cancer. In most cancers, the tumor cells can change their shapes to crawl through dense tissue, reach blood vessels, and spread to other sites. 

“We are now seeking to determine the physical limits of tumor cells’ deformability, and the forces that they must exert to push through a dense tissue,” Treado said.  Their work may lead to improvements in the ability to predict whether cancers will metastasize or not.