Metabolomics-enabled infrared spectral fingerprinting of allantoic fluid for chicken embryo sex determination

Allantoic fluid (AF) serves as an ideal matrix for early in ovo sex determination due to its accessibility and minimally invasive sampling. However, conventional methods typically rely on quantitative detection of single biomarkers (e.g., estrogen via ELISA), which is time-consuming and susceptible to matrix interference. To enable rapid in ovo sex identification during the critical window of sex differentiation, this study investigated the feasibility of AF metabolomic fingerprinting using attenuated total reflectance Fourier transform infrared (ATR‑FTIR) spectroscopy. Sex-specific metabolites in embryos at days 8–10 were profiled via liquid chromatography–tandem mass spectrometry (LC‑MS/MS), integrated with bioinformatics analysis, and validated by in vitro culture experiments. Additionally, an ultrafiltration‑based fractionation strategy was developed to enrich AF components, followed by spectral analysis and sex classification using multiple machine learning algorithms. Results indicated that metabolites exhibiting pronounced sexual dimorphism were predominantly detected in the low molecular-weight (MW) fraction of AF. Notably, ultrafiltration enrichment significantly improved spectral classification accuracy compared to untreated AF; the low‑MW fraction on day 8 yielded an area under the curve (AUC) of 0.93. Spectral regions at 1750–1500 cm −1 , 1459–1375 cm −1 , and 1200–1000 cm −1 showed significant correlation (| r | > 0.72, p < 0.05) with sex‑specific variations in amino acids, nucleic acids, and lipids, aligning with embryo‑specific metabolic patterns. These findings highlight the critical contribution of low‑MW AF components to spectral‑based sex discrimination. Consequently, the proposed ATR‑FTIR fingerprinting approach enables rapid acquisition of sex‑relevant metabolic information, offering a promising strategy for early, minimally invasive sex determination in poultry eggs.