Related Experiment Video
Updated: Jul 13, 2025
Enhanced Genetic Analysis of Single Human Bioparticles Recovered by Simplified Micromanipulation from Forensic ‘Touch DNA’ Evidence
Published on: March 9, 2015
Multi-omics integration strategy in the post-mortem interval of forensic science.
Jian Li1, Yan-Juan Wu1, Ming-Feng Liu1
1School of Forensic Medicine, Shanxi Medical University, No. 98, University Street, Wujinshan Town, Yuci District, Jinzhong City, Shanxi Province, 030604, PR China; Shanxi Key Laboratory of Forensic Medicine, Jinzhong, 030600, Shanxi, China.
Forensic scientists can now estimate the post-mortem interval (PMI) more accurately using a new multi-omics stacking model (MOSM). This AI-driven approach integrates various molecular data for improved PMI prediction in legal investigations.
Area of Science:
- Forensic Science
- Computational Biology
- Biochemistry
Background:
- Accurate post-mortem interval (PMI) estimation is crucial for forensic investigations.
- Current PMI estimation methods based on molecular markers have limitations in accuracy and comprehensiveness.
- There is a need for advanced techniques to improve the reliability of PMI determination.
Purpose of the Study:
- To develop and validate an innovative approach for accurate post-mortem interval (PMI) estimation.
- To integrate multi-omics data with artificial intelligence for enhanced PMI prediction.
- To address the limitations of existing methods in forensic science.
Main Methods:
- Development of the multi-omics stacking model (MOSM), integrating metabolomics, protein microarray electrophoresis, and Fourier Transform-Infrared Spectroscopy data.
- Application of machine learning models with distinct algorithmic principles within the MOSM framework.
- Utilizing multimolecular, multimarker, and multidimensional information for biological process description.
Main Results:
- The MOSM achieved a prediction accuracy of 0.93 for PMI.
- Demonstrated a generalized area under the receiver operating characteristic curve (AUC) of 0.98.
- Reported a minimum mean absolute error (MAE) of 0.07, indicating high precision.
Conclusions:
- The multi-omics stacking model (MOSM) significantly improves the accuracy and robustness of post-mortem interval estimation.
- Integration of diverse omics data and machine learning enhances the generalizability of PMI predictions.
- This AI-driven approach offers a more comprehensive and reliable tool for forensic applications.
