Spectroscopic characterization of human immunodeficiency virus type-1-infected plasma by principal component analysis and soft independent modeling of class analogy of visible and near-infrared spectra

Abstract

Previous research has shown that it may be possible to diagnose infections of human immunodeficiency virus type-1 (HIV-1) using plasma by a partial least squares regression analysis of visible and near-infrared (Vis-NIR) spectra. In this study, the features of plasma in HIV-1-infected and healthy individuals were further investigated by Vis-NIR spectroscopy using principal component analysis (PCA) and soft independent modeling of class analogy (SIMCA). Although the mean Vis-NIR spectra of 33 HIV-1-infected individuals and 15 healthy donors showed only slight differences, the two groups were respectively distinguished using a score plot of the first versus second or second versus third principal components, and by a Coomans plot. The PCA loadings were generally consistent with the discriminating power of the SIMCA, indicating specific changes in Vis-NIR spectra after HIV-1-infection. The specific pattern possibly indicates ROH and RNH2, which may constitute specific features of components in HIV-1-infected plasma.