A new algorithm could drive breakthroughs in understanding cancer, Alzheimer’s disease, and other potentially fatal conditions.<\/p>\n
Researchers from the University of Waterloo developed the machine learning algorithm, called RNovA, to detect changes in the proteins in human cells.<\/p>\n
Proteins do much of the work inside cells, and after they are made our bodies can chemically modify them in many ways. These changes, known as post-translational modifications, or PTMs, help regulate many cellular functions. Changes in PTMs have been linked to multiple serious diseases. \u00a0<\/p>\n
“Identifying PTMs in biological samples is expensive and technically challenging,” said Zeping Mao, a PhD candidate in computer science and lead author on the study. “This has traditionally been done in a lab using equipment like mass spectrometers. Using an algorithm is much faster and cheaper.<\/p>\n
Existing methods for identifying PTMs are most effective when researchers already know what they are looking for after consulting either a reference protein database, predefined list of modifications, or labeled training data.\u00a0<\/p>\n
“If a modification is rare, unexpected or missing from the database, existing methods can overlook it,” Mao said. “It’s like trying to solve a puzzle but only being able to see a few pieces.”\u00a0<\/p>\n
RNovA can quickly and accurately identify new PTMs that aren’t already in databases. This means the model can detect unexpected modifications without being retrained for each new PTM, and without starting from a predefined list of identified PTMs.\u00a0<\/p>\n
The team hopes that the discovery will lead to advances in diagnostics while also expanding the capabilities of machine learning in basic biological research.\u00a0<\/p>\n
“Expanding the PTM list may help researchers find new cellular modifications and new markers for cancer and other diseases,” Mao said. “It’s a very powerful tool that will help biologists to broaden their horizons.”\u00a0<\/p>\n