Deep Proteomics Network and Machine Learning Analysis of Human Cerebrospinal Fluid in Japanese Encephalitis Virus Infection
Résumé
Japanese encephalitis virus is a leading cause of neurologicalinfection in the Asia-Pacific region with no means of detection inmore remote areas. We aimed to test the hypothesis of a Japanese encephalitis(JE) protein signature in human cerebrospinal fluid (CSF) that couldbe harnessed in a rapid diagnostic test (RDT), contribute to understandingthe host response and predict outcome during infection. Liquid chromatographyand tandem mass spectrometry (LC-MS/MS), using extensive offlinefractionation and tandem mass tag labeling (TMT), enabled comparisonof the deep CSF proteome in JE vs other confirmed neurological infections(non-JE). Verification was performed using data-independent acquisition(DIA) LC-MS/MS. 5,070 proteins were identified, including 4,805human proteins and 265 pathogen proteins. Feature selection and predictivemodeling using TMT analysis of 147 patient samples enabled the developmentof a nine-protein JE diagnostic signature. This was tested using DIAanalysis of an independent group of 16 patient samples, demonstrating82% accuracy. Ultimately, validation in a larger group of patientsand different locations could help refine the list to 2-3 proteinsfor an RDT. The mass spectrometry proteomics data have been depositedto the ProteomeXchange Consortium via the PRIDE partner repositorywith the dataset identifier PXD034789 and 10.6019/PXD034789.