Spectroscopic Discrimination of Bone Samples from Various Species

Abstract

Determining the species of origin of skeletal remains is critical in a forensic and anthropologic context. However, there are very few methods that use a chemical approach to assist in this determination. In this study, Raman spectroscopy was used to discriminate bone samples originating from four different species (bovine, porcine, turkey and chicken). Spectra were obtained using a near infrared laser at 785-nm. All spectra were combined in a single matrix and proc- essed using partial least squares discriminate analysis (PLS-DA) with leave-one-out cross-validation. Three components were found to adequately describe the system. The first two components which contributed over 85% of spectral data was seen to completely separate the four species of origin in a two dimensional scores plot. A 95% confidence interval was draw around score points of each species class with very slight overlap. The first two components were seen to have large contributions from bioapatite and collagen, the main components of bone. This study serves as a preliminary investigation to evaluate the effectiveness of Raman spectroscopy to discriminate the species of origin of bone tissue.