Industry News
AI-Driven Proteomic Clock Deciphers Eye Aging and Disease Markers
A team of researchers has successfully mapped nearly 6,000 proteins across various cell types within the eye. Leveraging tiny drops of eye fluid routinely collected during surgeries, the scientists utilised an artificial intelligence (AI) model to construct a "proteomic clock" capable of predicting an individual's age based on their unique protein profile. The findings, reported in the journal Cell on October 19, unveil a potential nexus between ocular diseases, accelerated aging within specific cell types, and even early markers of Parkinson's disease.
The research, led by senior author Vinit Mahajan, a distinguished surgeon and professor of ophthalmology at Stanford University, sheds light on the dynamic relationship between anatomical changes and molecular processes occurring inside the eyes of patients. "What's amazing about the eye is we can look inside and see diseases happening in real time," Mahajan enthuses.
Sampling the eye for research purposes is particularly challenging due to its non-regenerative nature, akin to the brain. Traditional tissue biopsies risk irreparable damage. The team, therefore, opted for liquid biopsies, extracting fluid samples from the vicinity of cells or tissues of interest. However, limitations persisted, primarily in measuring large numbers of proteins within the minuscule volumes of fluid and in attributing proteins to specific cell types, crucial for effective disease diagnosis and treatment.
To surmount these challenges, Mahajan's team employed a high-resolution method to characterize proteins in 120 liquid biopsies from patients undergoing eye surgery. This led to the identification of an astonishing 5,953 proteins—ten times more than in previous studies. Using their proprietary software tool TEMPO, the researchers successfully traced each protein back to specific cell types.
The team then developed an AI machine learning model to predict the molecular age of the eye based on a subset of 26 proteins. Impressively, the model accurately gauged the age of healthy eyes while revealing significant molecular aging associated with diseases. In the case of diabetic retinopathy, aging escalated with disease progression, reaching up to 30 years' acceleration in individuals with severe forms.
Even more striking was the discovery of proteins linked to Parkinson's disease in the eye fluid. These proteins, typically identified postmortem, offer a potential avenue for earlier Parkinson's diagnoses and subsequent therapeutic monitoring—a breakthrough given the current diagnostic challenges.
The researchers propose that these findings indicate organ- or cell-specific aging, paving the way for advancements in precision medicine and clinical trial design. Julian Wolf, first author and ophthalmologist at Stanford University, highlights the potential for targeted anti-aging drugs in preventative precision medicine.
As the study progresses, the researchers plan to expand their sample size and include a broader spectrum of eye diseases. Moreover, they envision the applicability of their method in characterizing other challenging-to-sample tissues, such as cerebrospinal fluid for brain studies, synovial fluid for joint analysis, and urine for kidney examinations. This pioneering research opens new avenues for understanding aging and diseases at the molecular level, promising transformative impacts on clinical trials, drug selection, and outcomes.