Scientists are in the process of creating a deep learning network that can identify disease biomarkers with significantly higher precision.
Image Credit: University of Waterloo.
Researchers at the University of Waterloo’s Cheriton School of Computer Science have come up with a deep neural network that realizes 98% detection of peptide features in a dataset. This implies that researchers and medical practitioners have higher opportunities of identifying potential diseases by analyzing tissue samples.
There are already several techniques to identify diseases by investigating the protein structure of bio-samples. Computer programs have a major role in this process as they analyze a huge amount of data generated in such tests to identify specific markers of disease.
But existing programs are often inaccurate or can be limited by human error in their underlying functions. What we’ve done in our research is to create a deep neural network that achieves 98 percent detection of peptide features in a dataset. We’re working to make disease detection more accurate to provide healthcare practitioners with the best tools.
Fatema Tuz Zohora, PhD Researcher, Cheriton School of Computer Science, University of Waterloo
Peptides are the chains of amino acids constituting proteins in human tissue. These are the small chains that usually display the particular markers of disease. Better testing translates to a possibility of detecting diseases earlier and with higher precision.
Zohora and her colleagues have named their new deep learning network PointIso. It is a kind of artificial intelligence or machine learning trained on a huge database of existing sequences from bio-samples.
Other methods for disease biomarker detections usually have lots of parameters which have to be manually set by field experts. But our deep neural network learns the parameters itself, which is more accurate, and makes the disease biomarker discovery approach automated.
Fatema Tuz Zohora, PhD Researcher, Cheriton School of Computer Science, University of Waterloo
The new program is also exclusive because it is trained not only to look for a single kind of disease but to pinpoint the biomarkers related to a range of diseases, such as heart disease, cancer, and even COVID-19.
It’s applicable for any kind of disease biomarker discovery. And because it is essentially a pattern recognition model, it can be used for detection of any small objects within a large amount of data. There are so many applications for medicine and science; it’s exciting to see the possibilities opening up through this research and how it can help people.
Fatema Tuz Zohora, PhD Researcher, Cheriton School of Computer Science, University of Waterloo
Journal Reference:
Zohora, F. T., et al. (2021) Deep neural network for detecting arbitrary precision peptide features through attention-based segmentation. Scientific Reports. doi.org/10.1038/s41598-021-97669-7.