To help break that impasse, the researchers worked with a type of branched, tree-like polymer called a “bottlebrush.” Their inspiration came from biomolecules, whose various shapes determine their functions. Deshmukh said synthesizing them in the laboratory could lead to new medical treatments and other industry applications. But this can be difficult because polymers rapidly change shape depending on temperature and other factors. Without an efficient and accurate way of analyzing and predicting those changes, it is difficult to create a synthetic version.
Their new process uses a well-known deep-learning system, called a convolutional neural network, or CNN, to identify and predict similarities in shape and function in polymers—something that cannot be done without the aid of computers. could.
Deshmukh said that applying artificial intelligence to this polymer problem is “unprecedented because it shows the potential of deep learning methods in the field of soft materials.” “So, in theory, if we understand how the shapes are changing, hopefully we can control them.”
To prove that his method would work, Joshi ran 100 unique CNN models that taught the system to identify similarly sized bottlebrushes. The project was challenging, not only because painstaking work was required to teach the model what data and features to look for in polymers, but also because the researchers did not immediately know which features were relevant. They had to find out first.
Deshmukh said it took more than a year to develop the models. “Singh and Joshi did a great job of identifying the processing of the relevant data and then further refining it to allow the CNN model to get the correct information.”
“Most of the initial brainstorming was done by Dr Singh and Dr Deshmukh on the facilities to be used, which helped eliminate a lot of unfavorable options,” Joshi said. “This helped us to zero in on our current methodology, which I used to code and incorporate into my analysis algorithms.”
The results have been very promising, Joshi said, and the team hopes to expand the use of the technique to the growing field of glycomaterials — carbohydrate-based soft materials produced by every living organism.
These soft foods contain chains of sugars, called glycans, that play an important role in health and disease. Of the four building blocks of life – glycans, proteins, lipids and nucleic acids – glycans are the most complex and most challenging to understand. But CNN could accelerate progress in this area.
“So, just like we made these bottlebrush structures for synthetic polymers, there are a lot of architectures that can be built using glycomaterials and polymers like these glycans,” Deshmukh said.
Work on this paper was supported by the $23 million GlycoMIP projects, a multi-university partnership led by Virginia Tech and funded by the National Science Foundation. It was announced in 2020 to address scientific and technological barriers to glycomaterials research.
“We plan to help our partners design new types of glycomaterials that can be used for biomedical applications,” Deshmukh said. “It’s really exciting.”