Integrated analysis of restraint stress in rat serum using ATR-FTIR and Raman spectroscopy with Machine learning.

Journal: Spectrochimica acta. Part A, Molecular and biomolecular spectroscopy
Published Date:

Abstract

In forensic practice, accurately determining whether an individual has been subjected to prolonged restraint or assessing injuries resulting from restraint can be challenging. To address this, we explored a novel approach using attenuated total reflection Fourier transform infrared (ATR-FTIR) spectroscopy and Raman spectroscopy combined with machine learning to jointly identify restraint stress. We randomly assigned rats into three experimental groups: a restraint stress group (subjected to fasting and water deprivation), a control group (subjected to fasting and water deprivation without restraint), and a normal group. After collecting the serum spectra of the animals, a principal component analysis (PCA) model was established to explore the separation trends among the groups and to identify relevant characteristic peaks. Subsequently, a random forest (RF) model was established to compare the restraint stress group with the other two groups. The analysis identified key substances that indicated the presence of restraint stress: 1161 cm, 1167 cm (anti-symmetric C-O-C stretch) and 980 cm, 976 cm, 974 cm (-N+(CH3)3, antisymmetric stretch). And the RF model was used to compare the restraint stress groups at different time points, revealing substances that may help determine the duration of restraint stress: 1747 cm (ester carbonyl band), 1626 cm (β-pleated sheet), 1211 cm (Amide III, -N+(CH3)3, antisymmetric stretch), 1180 cm (phosphodiester), 1128 cm (-C-C-), 1024 cm (C-O stretching coupled with C-O bending) and 1389 cm, 1335 cm, 1321 cm (Trp, α helix, phospholipids), 710 cm (Polysaccharides), 1266 cm (Amide Ⅲ), 1015 cm (β-carotene). These findings suggest that ATR-FTIR and Raman spectroscopy together, when combined with machine learning, has significant potential as a powerful tool for analyzing and characterizing restraint stress, offering new insights and directions for future research in this area.

Authors

  • Zehua Fan
    Institute of Forensic Science, Fudan University, Shanghai 200032, People's Republic of China; College of Forensic Medicine, Hebei Key Laboratory of Forensic Medicine, Hebei Medical University, Shijiazhuang 050000, People's Republic of China; Department of Forensic Pathology, Institute of Forensic Science, Shanghai Key Laboratory of Forensic Medicine, Academy of Forensic Science, Shanghai 200063, People's Republic of China.
  • Chenyu Li
    Key Laboratory of Resources Biology and Biotechnology in Western China, Ministry of Education, Provincial Key Laboratory of Biotechnology of Shaanxi Province, College of Life Sciences, Northwest University, Xi'an 710069, China.
  • Qiran Sun
    Shanghai Key Laboratory of Forensic Medicine, Shanghai Forensic Service Platform, Academy of Forensic Science, Shanghai, China.
  • Yiwen Luo
    Shanghai Key Laboratory of Forensic Medicine, Shanghai Forensic Service Platform, Academy of Forensic Science, Shanghai, China.
  • Hancheng Lin
    Shanghai Key Laboratory of Forensic Medicine, Shanghai Forensic Service Platform, Academy of Forensic Science, Shanghai, China.
  • Bin Cong
    College of Forensic Medicine, Hebei Medical University, Hebei Key Laboratory of Forensic Medicine, Shijiazhuang 050017, China. Electronic address: hbydbincong@126.com.
  • Ping Huang
    Division of HIV/AIDS Prevention, Centers for Disease Control and Prevention, Atlanta, GA, USA.