Scholarlink

<p>Estimating the postmortem interval (PMI) and identifying decomposition sites are important challenges in forensic science. The soil microbiome has shown potential in both applications. This study used pig models to simulate human decomposition and analyzed soil microbial succession over both short (0–11 days) and mid-to-long-term (up to 10 months) intervals to develop a PMI estimation model, while simultaneously comparing control and decomposition soils to assist in identifying potential corpse deposition sites. We built KNN (K-nearest neighbors) models at the genus level, which achieved high performance in short-term PMI estimation. Furthermore, a Linear Discriminant Analysis (LDA) model demonstrated robust performance in long-term PMI estimation, relying primarily on animal-associated microbial genera. Additionally, the KNN machine learning model effectively distinguished soils impacted by cadavers. This study provides a promising tool for estimating PMI and identifying potential body deposition sites.&#xa0;</p>

错误：搜索内容不能为空，请输入英文关键词
错误：关键词超出字数限制，请精简
高级检索

Soil microbial succession for forensic estimation of postmortem interval and decomposition site identification

  • 期刊论文

  • 发表日期：

    2026年4月23日

查看全文
  • Yang Xia,
  • Sile Chen,
  • Zhe Deng,
  • Chengxin Ye,
  • Hai Wu,
  • Yuanxing Wang,
  • Shaojiang Guo,
  • Nan Guo,
  • Junbo Yang,
  • Zhongxiu Tao,
  • Xinquan Xiong,
  • Yadong Guo,
  • Yanjie Shang

摘要

Estimating the postmortem interval (PMI) and identifying decomposition sites are important challenges in forensic science. The soil microbiome has shown potential in both applications. This study used pig models to simulate human decomposition and analyzed soil microbial succession over both short (0–11 days) and mid-to-long-term (up to 10 months) intervals to develop a PMI estimation model, while simultaneously comparing control and decomposition soils to assist in identifying potential corpse deposition sites. We built KNN (K-nearest neighbors) models at the genus level, which achieved high performance in short-term PMI estimation. Furthermore, a Linear Discriminant Analysis (LDA) model demonstrated robust performance in long-term PMI estimation, relying primarily on animal-associated microbial genera. Additionally, the KNN machine learning model effectively distinguished soils impacted by cadavers. This study provides a promising tool for estimating PMI and identifying potential body deposition sites.