Direct ultra-sensitive detection of pathogen biomarkers in blood could provide a universal strategy for diagnosis of bacterial infections, which remain a leading cause of morbidity and mortality in many areas of the world. Many factors complicate diagnosis, including the presence of multiple co-infections in a given patient, and lack of infrastructure in rural settings. In some pediatric patients, such as those in areas with poor resources, an additional challenge exists with low sample volumes due to age and other health factors such as anemia and dehydration. Our team is working on developing novel diagnostic assays, with a waveguide-based biosensor platform, to rapidly and specifically identify pathogen biomarkers from small samples of serum or plasma, allowing for the timely and sensitive diagnosis of infection at the point of care. In addition to the platform, we have developed novel membrane insertion and lipoprotein capture assay methods, to capture lipidated pathogen biomarkers in aqueous blood, by virtue of their interactions with host lipoprotein carriers. Herein, we demonstrate our efforts to adapt the lipoprotein capture assay for the detection of small concentrations of pathogen-secreted lipopolysaccharides in aqueous blood, with the ultimate aim of diagnosing Gram-negative infections effectively.
Diagrams of cellular processes present a clean, deterministic view of how biomolecules regulate the processes of life. Attempts to construct reaction networks which are true to the microscopic complexity of the system are intractable when only a few proteins are included. We argue here that several layers of microscopic modeling are needed to characterize the fluctuations, or noise, of biochemical systems and that this is necessary to develop predictive models of cellular processes. Our arguments are illustrated with the specific examples of myoglobin and protein kinases.