When Netflix recommends us to watch a series or Facebook suggests a person to send us a friend request, they use powerful machine learning algorithms that allow them to make predictions about what we will do next. Now, a team of researchers has used this artificial intelligence technique to decipher the biological language of cancer and neurodegenerative diseases such as Alzheimer's
Proteins that destroy nerve cells and cause Alzheimer's
Many key functions of the human body are performed by proteins, large and complex molecules that do most of the work in cells and are necessary for the structure, function and regulation of tissues and organs; Antibodies, for example, are a protein whose function is to protect the body.
In the case of Alzheimer's disease, proteins undergo an alteration that causes them to form groups and destroy healthy nerve cells. When the brain is healthy it has a kind of quality control system capable of effectively removing accumulations of potentially dangerous proteins called aggregates.
Bringing natural language processing technology to investigating the molecular origins of protein malfunction is vital if we are to be able to correct grammatical errors within cells that cause disease.
Learn the "rules of disease language"
Dr. Kadi Liis Saar -researcher at St. John's College, Cambridge University- used machine learning technology similar to platforms such as Netflix, Facebook or Amazon to develop a large-scale model of language that would allow us to observe what occurs when the proteins of the body are altered and this causes the appearance of a disease. The expert explained that “the human body houses thousands and thousands of proteins and scientists still do not know the function of many of them. We asked a neural network-based language model to learn the language of proteins. "
“We specifically asked the program to learn the language of the shape-shifting biomolecular condensates (protein droplets found in cells) that scientists really need to understand in order to decipher the language of biological function and malfunction that cause cancer and neurodegenerative diseases such as Alzheimer's. We found that it could learn, without being explicitly told, what scientists have already discovered about the language of proteins during decades of research, ”said Liis Saar.
According to the study authors, their findings could be used in the future to "correct the 'grammatical errors' within cells that cause disease," and could contribute to new discoveries that transcend the human brain's ability to understand why the body gets sick.