EUVIMEDEUVIMED entdecken
Crossref

Explainable artificial intelligence in forensic DNA analysis: Alleles identification in challenging electropherograms using supervised machine learning methods

Mengyu Tan, Yuxuan Tan, Haoyan Jiang, Jiaming Xue, Qiushuo Wu, Yazi Zheng, Guihong Liu, Yuanyuan Xiao, Meili Lv, Miao Liao, Lin Zhang, Shengqiu Qu und 1 weitere

Forensic Science International: Genetics · 2025-06 · 78 · 103289

DOI: 10.1016/j.fsigen.2025.103289 ↗

Kurz gefasst

Fachartikel in der Zeitschrift Forensic Science International: Genetics, 2025-06. Verfasst von Mengyu Tan und 12 weiteren Autor:innen. Beteiligte Einrichtung: Department of Forensic Genetics, West China School of Basic Medical Sciences and Forensic Medicine, Sichuan University, Chengdu, China. Hauptthemen laut MEDLINE: Supervised Machine Learning, DNA Fingerprinting, Alleles, Artificial Intelligence. Die Arbeit wurde bisher 10-mal zitiert (Crossref) und nennt 53 Literaturangaben. Gefördert von: National Natural Science Foundation of China, Sichuan Provincial Science and Technology Support Program.

AbstractSchlagwörterBibliografische AngabenZitierenVerwandte Studien

Abstract (Originalsprache)

Challenging samples in capillary electrophoresis (CE)-based short tandem repeat (STR) analysis often produce artefactual signals that cannot be completely filtered out by expert electropherogram (EPG) reading systems, complicating allele interpretation. Previous studies have demonstrated the potential of artificial intelligence (AI) to address this issue by accurately distinguishing allele signals from artefacts in EPGs. Traditional machine learning models offer significant advantages in enhancing the interpretability and transparency of AI models used in DNA analysis, particularly in criminal investigations and legal contexts. In this study, five traditional machine learning algorithms were employed to train and construct models using EPG signal datasets from single-source low-template EPGs, mixture EPGs, and combined datasets. Performance evaluation and validation with additional datasets demonstrated the feasibility of these models in improving the reportability of potential information in EPGs. However, further optimization is needed for mixture EPGs to enhance classification accuracy. Implementing Receiver Operating Characteristic (ROC) curve analysis and prediction probability thresholds effectively reduced false positive classifications. Additionally, a user-friendly platform was developed for EPG signal classification based on machine learning and ensemble learning, allowing for the classification of any signal datasets using traditional machine learning models and combining the prediction results of multiple models. This platform will provide analysts with more optimal and robust results. This study shows that machine-learning-based EPG signal classification models can significantly enhance the efficiency of sample analysis and interpretation, providing a solid foundation for future research.

Schlagwörter

MeSH-Schlagwörter (MEDLINE)

Supervised Machine LearningDNA FingerprintingAllelesArtificial IntelligenceMachine LearningHumansMicrosatellite RepeatsElectrophoresis, CapillaryROC CurveAlgorithmsForensic Genetics

Stichwörter der Autor:innen

Alleles identificationCapillary electrophoresisElectropherogramMachine learningShort tandem repeat analysis

Autor:innen und Institutionen

  • Department of Forensic Genetics, West China School of Basic Medical Sciences and Forensic Medicine, Sichuan University, Chengdu, China.
  • Department of Immunology, West China School of Basic Medical Sciences and Forensic Medicine, Sichuan University, Chengdu, Sichuan 610041, China.
  • Department of Forensic Genetics, West China School of Basic Medical Sciences and Forensic Medicine, Sichuan University, Chengdu, China. Electronic address: qushengqiu@scu.edu.cn.
  • Department of Forensic Genetics, West China School of Basic Medical Sciences and Forensic Medicine, Sichuan University, Chengdu, China. Electronic address: liangweibo@scu.edu.cn.

Förderung

  • National Natural Science Foundation of China
  • Sichuan Provincial Science and Technology Support Program

Zitieren

APA

Tan, M., Tan, Y., Jiang, H., Xue, J., Wu, Q., Zheng, Y., Liu, G., Xiao, Y., Lv, M., Liao, M., Zhang, L., Qu, S., & Liang, W. (2025). Explainable artificial intelligence in forensic DNA analysis: Alleles identification in challenging electropherograms using supervised machine learning methods. Forensic Science International: Genetics, 78, 103289. https://doi.org/10.1016/j.fsigen.2025.103289

Vancouver

Tan M, Tan Y, Jiang H, Xue J, Wu Q, Zheng Y, et al. Explainable artificial intelligence in forensic DNA analysis: Alleles identification in challenging electropherograms using supervised machine learning methods. Forensic Science International: Genetics. 2025;78:103289. doi:10.1016/j.fsigen.2025.103289

BibTeX
@article{Tan2025,
  author = {Tan, Mengyu and Tan, Yuxuan and Jiang, Haoyan and Xue, Jiaming and Wu, Qiushuo and Zheng, Yazi and Liu, Guihong and Xiao, Yuanyuan and Lv, Meili and Liao, Miao and Zhang, Lin and Qu, Shengqiu and Liang, Weibo},
  title = {Explainable artificial intelligence in forensic DNA analysis: Alleles identification in challenging electropherograms using supervised machine learning methods},
  journal = {Forensic Science International: Genetics},
  year = {2025},
  volume = {78},
  pages = {103289},
  doi = {10.1016/j.fsigen.2025.103289}
}
Kostenlos starten →