Scientists Come up with New AI Tool to Study Protein Dynamics inside Human Cells
It is well known that proteins in human cells control a range of biological processes. Understanding the working of protein in the cells is the most important part of fundamental biology. The main tool for studying protein motion is known as single-molecule Forster Resonance Energy Transfer (smFRET).
A recent study published in eLife has revealed that scientists have come up with an artificial intelligence tool to analyse how proteins move and interact. This tool is believed to be faster and more accurate in nature. This tool has been made freely available and has been dramatically speeding up the study of protein dynamics.
Explaining the functioning of the tool, the research mentions that two or more parts of the molecule are labelled with different fluorescent tags. When the two of them come in close proximity, then the change in fluorescence can be detected. With the detection done by the microscope, one can visualise and measure down the nanometre level of the movement of proteins.
Johannes Thomsen, the lead author of the study and a research assistant at the University of Copenhagen, Denmark, mentioned some of the challenges that one would face while using this technology. He asserted that the challenge regarding the analysis of the huge data still persists.
The lead author said, "Some of the challenges with single-molecule Forster Resonance Energy Transfer include the very large data that are produced, and the steps that researchers need to take to process the images before analysis."
Emphasising upon how this tool is beneficial for research and how it is helpful for understanding large datasets without the need for human intervention, he said, "Machine learning technologies, especially deep neural networks, have significantly improved our ability to understand large datasets without the need for human intervention. We wanted to see whether employing these technologies to single-molecule Forster Resonance Energy Transfer data would allow automated, fast characterisation of protein motions, independently of human experts."