CircBioCare partner Covaris and researchers from Maastricht University have published an application note showing that Adaptive Focused Acoustics (AFA) technology detects more proteins, including low-abundance ones, in knee fat tissue from patients with osteoarthritis.
Finding reliable knee osteoarthritis biomarkers is one of the main challenges in joint health research. Osteoarthritis is very common, yet there are still no validated biomarkers that can detect it before the first symptoms appear. A new application note from Covaris, a partner in the CircBioCare project, shows how an automated sample preparation method can help close this gap.
Why early diagnosis of osteoarthritis needs better proteomics
Proteomics, the large-scale analysis of the proteins in a tissue, is one of the most promising ways to find new biomarkers. However, traditional protein extraction methods are manual and hard to reproduce, and they often miss proteins present in very small amounts. These low-abundance proteins may be the ones that reveal most about how the disease starts and progresses.
The application note, “Optimizing Automated Sample Preparation with Adaptive Focused Acoustics (AFA) for Proteomics of Human Adipose Tissue”, published in October 2026, tackles this problem directly.
Studying the infrapatellar fat pad
The study focuses on Hoffa’s fat pad, also known as the infrapatellar fat pad, a tissue inside the knee that is thought to play a key role in the development of osteoarthritis. The samples came from patients who had knee surgery at Maastricht University Medical Center+.
The team compared two ways of extracting proteins from the tissue: the conventional method, with mechanical homogenization followed by freeze-thaw cycles, and the Covaris AFA technology, with the R230 Focused-ultrasonicator, which applies non-contact, focused acoustic energy to samples in sealed vessels, with precise temperature control.
More proteins identified, including low-abundance ones
The conventional method recovered more total protein, but AFA identified more distinct proteins. It detected an average of 1,623 proteins, compared with 1,377–1,480 for the conventional method, depending on the extraction buffer.
The difference was even larger for low-abundance proteins. AFA detected 380–412 of them, compared with 180–296 for traditional homogenization.
The study also showed that the method works consistently, even when the settings are changed. The extraction buffer used to extract the proteins made the biggest difference. And because the whole process is automated, it reduces errors and gives more reliable results when many patient samples are analysed.
The authors note that more research is needed to confirm the role of these low-abundance proteins.
Read the full application note here.
Authors: Magdalena Perczak, Luojiao Huang, Pieter Emans and Berta Cillero-Pastor (Maastricht University); Sameer Vasantgadkar, Debadeep Bhattacharyya and Akshay Dhingra (Covaris).



