Research project Collaborative project funded by TAIGA and WADA.
Our research group is at the forefront of developing a detection method for Cryo-RBC doping. Still, we face significant challenges due to the complex RBC proteome, which contains approximately 5,000–10,000 proteins and 30,000–50,000 peptides, among which only a few may serve as viable biomarkers. Unlike traditional clinical biomarkers for disease, the Cryo-RBC transfusion model presents several advantages for method development, primarily due to the well-controlled intervention that clearly distinguishes between doped and undoped states.
Doping is a prohibited method of enhanced performance, regulated by the World Anti-Doping Agency (WADA) and depends on robust detection methods. Blood doping - especially the practice of autologous blood doping using freeze-stored red blood cells (Cryo-RBCs) has been a prevalent strategy since the 1970s, and still today, without a reliable detection method.
Developing techniques that isolate specific cell populations without relying on antibodies or exogenous labels is crucial for pre-omics cell sorting in any context. Such pre-sorting methods not only simplify the omics data by reducing complexity but also enhance the concentration of potential biomarkers by selectively analysing cells with properties detected through artificial intelligence (AI).
The pilot project will validate interpretable ML approaches with the specific purpose of pre-omics cell sorting. Its outcome will provide us further insights into the ML framework and thus advance the state of the art to craft robust funding proposals in the coming years. In general, beyond conventional image generation and processing that relies on humans' limited knowledge and capability, we anticipate that applying ML methods directly to compressive modalities will have broad applicability for the real-time application of high quantity and high-dimensional data.
