AI Predicts Molecular Age & Eye Fluid Diseases

Recent study showed that the eye fluid contains information about molecular age and state of health of the eye. Scientists analyzed microscopic drops of liquid, usually removed during surgery by eye, and found almost 6,000 proteins belonging to various cellular types of eyes.

As part of the study, an artificial intelligence model was also developed, which can evaluate the age of a person based on his protein profile. The work was published in the magazine “Cell” on October 19.

The leading author of the study, Dr. Blames Mahajan, professor of ophthalmology at Stanford University, emphasized the uniqueness of the eye as an organ that allows observing the development of diseases. He stated that the goal was to connect visible changes in the eye with molecular changes in the liquid.

To analyze the fluid of the eye, the researchers used the method of liquid biopsy, which allowed them to detect 10 times more proteins than in previous studies. Using the TEMPO software tool, they were able to determine the cellular type of each protein.

Using a machine learning model, scientists have created a “protein watch” that can predict the molecular age of the eye based on 26 proteins. The model showed that diseases such as diabetic retinopathy and uveitis cause accelerated aging of certain cellular types.

The study also revealed the presence of proteins associated with Parkinson’s disease in the eye fluid. These proteins are usually found only after death, which complicates the diagnosis of Parkinson’s disease. Scientists proposed using the screening of these proteins in the eyes of the eye for early diagnosis and monitoring the progression of the disease.

The authors emphasized that their results may have consequences for accurate medicine and clinical trials. They also stated the intention to analyze samples from a larger number of patients and diseases of the eye.

/Reports, release notes, official announcements.