Minimal Invasive Gender Determination of Birds by Means of UV-Resonance Raman Spectroscopy

• Published in: Analytical chemistry (Washington) Vol. 80; no. 4; pp. 1080 - 1086
• Main Authors: Harz, M, Krause, M, Bartels, T, Cramer, K, Rösch, P, Popp, J
• Format: Journal Article
• Language: English
• Published: Washington, DC American Chemical Society 15-02-2008
• Subjects: Analytical chemistry; Animals; Birds; Breeding of animals; Chemistry; Chickens; Cluster analysis; DNA fingerprints; DNA-Binding Proteins - metabolism; Exact sciences and technology; Feathers - physiology; Female; Gender; Male; Principal Component Analysis; Proteins; Sensitivity and Specificity; Sex Determination Analysis - methods; Sex Differentiation - physiology; Spectrometric and optical methods; Spectrophotometry, Ultraviolet - methods; Spectrum analysis; Spectrum Analysis, Raman - methods; Cluster analysis; Vertebrata; Sample preparation; Investigation method; DNA; Raman spectrometry; Aves; Sampling; Ultraviolet spectrometry; Protein; Blood; Principal component analysis
• ISSN: 0003-2700; 1520-6882
• DOI: 10.1021/ac702043q

The identification of avian gender is important for prosperous breeding of birds. Since birds do not possess external genital organs, endoscopic investigations, blood analysis, and molecular biological methods are applied to determine the gender in monomorphic species. However, anesthesia and blood sampling impose stress on the examined bird and should be avoided in terms of animal protection. Here we report on the application of UV-resonance Raman spectroscopy as a minimal invasive method for gender determination of birds via an evaluation of feather pulp samples. Sample preparation for this investigation method is simple and facilitates a quick and easy analysis. The UV-resonance Raman spectra of the feather pulp sample extracts are dominated by DNA and protein signals. The different DNA content in male and female chicken allows for gender differentiation via its characteristic Raman fingerprint. The classification either to male or female chicken is ideally accomplished by support vector machines due to the fact that no unknown classes are involved. Recognition rates of about 95% were compared to less effective results of the unsupervised hierarchical cluster analysis. Within the scope of our investigations, principal component analysis was also applied to determine the important spectral regions for the classification of chicken's feather pulp samples.