Using silicon-based biosensors to detect host protein signatures capable of differentiating between bacterial and viral infection at point-of-care

The World Health Organisation has predicted that the rise of antimicrobial resistance (AMR) will claim up to 10 million lives by the year 2050[1]. One of the key factors driving AMR emergence is the lack of a rapid diagnostic device capable of differentiating between bacterial and viral infection. A bacterial culture test is the current "gold standard", however it presents with various draw backs such as long time-to-positive results and risk of sample contamination[2]. This leads to clinicians either delaying antibiotic treatment as they wait for confirmation or overprescribing broad spectrum antibiotics in cases of severe infection. It is then unsurprising to learn that around 50% of antibiotics are inappropriately prescribed[3].
The ideal diagnostic device would be rapid and available at point-of-care with high sensitivity and specificity, however challenges with current diagnostics make these characteristics difficult to achieve. Primarily, they mostly rely on optical-based methods of detection which usually require bulky and expensive laboratory equipment and so often makes these assays unavailable at point-of-care[4]. Additionally, direct detection of the causative pathogen may not be possible as some pathogens are incredibly invasive or inaccessible, especially in cases of tuberculosis. Finally, confirmation of a bacterial infection does not rule out a causative viral infection and vice versa which is a very common dynamic in upper respiratory infections[5].
Host protein signatures could bypass all these challenges and various host protein signatures have already been cited in literature. Oved et al. (2015) discovered a 3 host protein signature composed of C-reactive protein (CRP), TNF-related apoptosis-inducing ligand (TRAIL) and Interferon gamma-induced protein-10 (IP-10) which is able to accurately differentiate between bacterial and viral infection[6]. However, it is estimated that over 90% of discovered host biomarkers have not yet been applied successfully at point-of-care[7].
Combining host protein signatures with already leveraged complementary metal-oxide semiconductor technology (CMOS) (the leading technology used to manufacture everyday electronics) would make such devices incredibly cheap and scalable[8]. CMOS technology is also compatible with ion sensitive field effect transistors (ISFETs), which can be used to detect changes in surface charge due to antigen-antibody binding and hence detect and quantify host proteins in real time.

Research questions:
-What specific host protein signature is able to differentiate between bacterial and viral infection with high sensitivity and specificity?
-How does the sensitivity and specificity of the immunoassay immobilised on-chip compare to the "gold standard" protein detection technique?
-What techniques are required to immobilise primary antibodies onto an array of ISFET sensors?
-What is the on-chip limit of detection for that specific protein signature?

Aims:
-This project aims to develop an immunoassay specific to a host protein signature capable of differentiation between bacterial and viral infection.
-The specific immunoassay will then be translated onto a silicon-based biosensor and the on-chip limit of detection will be established.
-Finally, the assay will be validated against clinical samples and used to provide quantitative data around the infection state of the patient.

References
[1] M. E. A. de Kraker, A. J. Stewardson, and S. Harbarth, "Will 10 Million People Die a Year due to Antimicrobial Resistance by 2050?," PLOS Medicine, vol. 13, no. 11, p. e1002184, Nov. 2016, doi: 10.1371/JOURNAL.PMED.1002184.
[2] J. C. Craig et al., "The accuracy of clinical symptoms and signs for the diagnosis of serious bacterial infection in young febrile children: prospective cohort study of 15 781 febrile illnesses," BMJ (Clinical research ed.), vol. 340, no. 7754, p. 1015, May 2010, doi: 10.1136/BMJ.C1594.
[3] C. Giuliano, C. R. Patel, and