Abstract:
Deaths due to Human Immunodeficiency Virus (HIV) related diseases are still very high, especially in Africa. This is mainly attributed to poor diagnosis and late detection of the virus at times when the virus has greatly multiplied within the human body. In an attempt to come up with early detection mechanisms of the virus, several investigations have been carried out. The current detection mechanisms are less rapid with some being. Here we report on the potential of Raman spectroscopy as an alternative optical method for early and rapid detection of HIV-1 p24 in plasma and HIV-1 in whole human blood and in plasma. Raman spectra of HIV-1 p24 antigen, HIV-1 positive (HIV+) and HIV-1 negative (HIV-) blood and plasma were analyzed in the fingerprint spectral range 400-1800 cm-1 after 785 nm excitation. The Raman spectra of HIV1-p24 antigen displayed characteristic Raman bands centered at 891, 967, 1004, 1080, 1170, 1200, 1270, 1335, 1449, 1486, 1609, 1656 and 1738 cm-1. Unique Raman bands were observed at around wavenumbers 928 cm-1and 1658 cm-1 for HIV- and at 1270 cm-1 and 1446 cm-1 for HIV+ plasma samples. The prominent Raman band at 1270 cm-1 and 1446 cm-1 in HIV+ samples were ascribed to vibrational state of amide III of α-helix and C-H bond bending vibrational mode in proteins and lipids components of plasma respectively. For multivariate analysis; Principal Component Analysis (PCA) and Artificial Neural Network (ANN) were used. PCA was able to differentiate clearly Raman spectral data from HIV- and HIV+ plasma samples with sensitivity of 100% and specificity of 89.28% as well as sensitivity of 83.3% and specificity of 100% for whole blood samples. Classification based on ANN applied on the entire spectral range of data sets gave high correlation coefficient R = 0.99798 and R = 0.99997 for HIV+ blood and plasma samples respectively. Similarly, quantification also resulted in a linear model with R2 = 0.9457 and R2 = 0.9959 based on unique HIV+ peak intensity centered at 1270 cm-1 and 1446 cm-1
