Status of Forensic Microbiology Research Based on Machine Learning

doi:10.3969/j.issn.1671-2072.2024.03.005

Abstract

Abstract: Microorganisms are widely present in nature and in human body. Their diverse community distribution provides new possibilities for solving forensic science challenges. As one of the main tools in artificial intelligence，machine learning plays a crucial role in data identification，processing and analysis，providing new ideas and methods for forensic microbiology research. This paper introduces various machine learning（ML） algorithms and related models，including supervised learning，unsupervised learning and deep learning，which have been widely used in recent forensic microbiology studies，and summarizes relevant studies on individual identification，postmortem interval inference，geographic location traceability. In addition，this paper systematically analyses challenges faced by machine learning algorithms in standardization，database establishment and evidence interpretation in forensic microbiology research，and discusses its application in forensic microbiology.

Key words: forensic science, microbiology, artificial intelligence（AI）, machine learning（ML） algorithm

摘要： 微生物广泛存在于自然界和人体，其群落分布差异性为解决法医学相关问题提供了很多新的可能性。作为实现人工智能的主要方法之一，机器学习拥有强大的识别、处理和分析数据能力，为法医微生物研究提供了新思路和新方法。通过介绍近年来在法医微生物研究中使用较多的无监督学习、监督学习和深度学习等机器学习算法，并对其在个体识别、死亡时间推断、地理位置溯源等领域的探索性研究进行总结。此外，还系统剖析了机器学习算法在法医微生物研究中标准化流程、数据库建立和证据解释等方面存在的问题，并对其在法医微生物中的应用进行展望。

关键词: 法医学, 微生物, 人工智能, 机器学习算法

CLC Number:

ZHANG Liwei, WANG Tian, YU Daijing, ZHANG Jun, YAN Jiangwei. Status of Forensic Microbiology Research Based on Machine Learning[J]. Chinese Journal of Forensic Sciences, 2024(3): 32-40.

张立为, 王甜, 余代静, 张君, 严江伟. 基于机器学习的法医微生物研究现状[J]. 中国司法鉴定, 2024(3): 32-40.

References

[ 1 ] MARCHESI J R，RAVEL J.The vocabulary of microbiome research： A proposal[J]. Microbiome，2015，30：3：31.
[ 2 ] BUDOWLE B，SCHUTZER S E，EINSELN A，et al. Public health. Building microbial forensics as a response to bioterrorism[J]. Science， 2003，301（5641）：1852-1853.
[ 3 ] TURNBAUGH P J， LEY R E， HAMADY M，et al. The human microbiome project[J].Nature，2007，449（7164）：804-810.
[ 4 ] FIERER N，LAUBER C L，ZHOU N，et al. Forensic identification using skin bacterial communities[J].Proceedings of the National Academy of Sciences of the United States of America， 2010，107（14）：6477-6481.
[ 5 ] GILBERT J A， BLASER M J， CAPORASO J G， et al. Current understanding of the human microbiome[J].Natural Medicine， 2018， 24（4）： 392-400.
[ 6 ] KULESHOV V， JIANG C， ZHOU W， et al. Synthetic long-read sequencing reveals intraspecies diversity in the human microbiome[J].Nature Biotechnology，2016，34（1）：64-69.
[ 7 ] HANDELSMAN J， RONDON M R， BRADY S F，et al. Molecular biological access to the chemistry of unknown soil microbes： A new frontier for natural products[J]. Chemistry & Biology，1998，5（10）：R245-R249.
[ 8 ] QUINCE C， WALKER A W， SIMPSON J T， et al. Shotgun metagenomics， from sampling to analysis [J]. Nature Biotechnology，2017，35（9）：833-844.
[ 9 ] METCALF J L， XU Z Z， WEISS S， et al. Microbial community assembly and metabolic function during mammalian corpse decomposition[J]. Science，2016，351（6269）：158-162.
[10] RAVI R K， WALTON K， KHOSROHEIDARI M. MiSeq： A next generation sequencing platform for genomic analysis[J]. Methods in Molecular Biology，2018，1706：223-232.
[11] BHARTI R， GRIMM D G. Current challenges and best-practice protocols for microbiome analysis[J]. Briefings in Bioinformatics，2021， 22（1）：178-193.
[12] JOHNSON J S， SPAKOWICZ D J， HONG B Y， et al. Evaluation of 16S rRNA gene sequencing for species and strain-level microbiome analysis[J]. Nature Communications，2019，10（1）：5029.
[13] PRODAN A， TREMAROLI V， BROLIN H， et al. Comparing bioinformatic pipelines for microbial 16S rRNA amplicon sequencing[J]. PLoS One，2020，15（1）：e0227434.
[14] JORDAN M I， MITCHELL T M. Machine learning： Trends， perspectives， and prospects[J]. Science，2015，349（6245）：255-260.
[15] ASNICAR F， THOMAS A M， PASSERINI A， et al. Machine learning for microbiologists[J]. Nature Reviews Microbiology， 2024，22（4）：191-205.
[16] BEN SALEM K， BEN ABDELAZIZ A. Principal component analysis（PCA） [J]. La Tunisie Medicale，2021，99（4）：383-389.
[17] DOBAY A， HAAS C， FUCILE G， et al. Microbiome-based body fluid identification of samples exposed to indoor conditions[J]. Forensic Science International Genetics，2019，40：105-113.
[18] ZHANG J， WANG M C， QI X Q， et al. Predicting the postmortem interval of burial cadavers based on microbial community succession[J]. Forensic Science International Genetics，2021，52：102488.
[19] BECKER T， ROUSSEAU A J， GEUBBELMANS M， et al. Decision trees and random forests[J]. American Journal of Orthodontics and Dentofacial Orthopedics，2023，164（6）：894-897.
[20] YANG J， TSUKIMI T， YOSHIKAWA M， et al. Cutibacterium acnes （Propionibacterium acnes） 16S rRNA genotyping of microbial samples from possessions contributes to owner identification[J]. mSystems， 2019，4（6）：e00594-e00619.
[21] METCALF J L， WEGENER P L， GONZA-LEZ A， et al. A microbial clock provides an accurate estimate of the postmortem interval in a mouse model system[J]. Elife，2013，15：2：e01104.
[22] DEEL H， EMMONS A L， KIELY J， et al. A pilot study of microbial succession in human rib skeletal remains during terrestrial decomposition[J]. mSphere，2021，6（4）：e0045521.
[23] WANG J Q， CHENG X J， ZHANG J， et al. Estimating the time since deposition （TsD） in saliva stains using temporal changes in microbial markers [J]. Forensic Science Internatioanal Genetics，2022，60：102747.
[24] MetaSUB International Consortium.The Metagenomics and Metadesign of the Subways and Urban Biomes （MetaSUB） International Consortium inaugural meeting report[J]. Microbiome，2016，4（1）：24.
[25] RYAN F J. Application of machine learning techniques for creating urban microbial fingerprints[J]. Biology Direct，2019，14（1）：13.
[26] ZHANG J， YU D J， WANG Y Y， et al. Tracing recent outdoor geolocation by analyzing microbiota from shoe soles and shoeprints even after indoor walking[J]. Forensic Science International Genetics，2023， 65：102869.
[27] MARILL K A. Advanced statistics：Linear regression， part II：Multiple linear regression[J]. Academic Emergency Medicine，2004， 11（1）：94-102.
[28] STOLTZFUS J C. Logistic regression：A brief primer[J]. Academic Emergency Medicine，2011，18（10）：1099-1104.
[29] HUANG L H， XU C Q， YANG W X， et al. A machine learning framework to determine geolocations from metagenomic profiling[J]. Biology Direct，2020，15（1）：27.
[30] OH J， BYRD A L， PARK M， et al. Temporal stability of the human skin microbiome[J]. Cell，2016，165（4）：854-866.
[31] SCHMEDES S E， WOERNER A E， BUDOWLE B. Forensic human identification using skin microbiomes[J]. Applied and Environmental Microbiology，2017，83（22）：e01672-e01717.
[32] SCHMEDES S E， WOERNER A E， NOVROSKI N M M， et al. Targeted sequencing of clade-specific markers from skin microbiomes for forensic human identification[J]. Forensic Science International Genetics，2018， 32：50-61.
[33] DÍEZ LÓPEZ C， KAYSER M， VIDAKI A. Estimating the time since deposition of saliva stains with a targeted bacterial DNA approach： A proof-of-principle study[J]. Frontiers Microbiology，2021，12： 647933.
[34] XANTHOPOULOS P， PARDALOS P M， TRAFALIS T B， et al. Linear discriminant analysis[J]. Robust Data Mining，2013，2：27-33.
[35] MEI S， ZHAO M， LIU Y， et al. Evaluations and comparisons of microbial diversities in four types of body fluids based on two 16S rRNA gene sequencing methods[J]. Forensic Science International，2022，331： 111128.
[36] HEARST M A， DUMAIS S T， OSUNA E， et al. Support vector machines[J]. IEEE Intelligent Systems and Their Applications，1998，13（4）：18-28.
[37] SHERIER A J， WOERNER A E， BUDOWLE B. Population informative markers selected using Wright’s fixation index and machine learning improves human identification using the skin microbiome[J]. Applied and Environmental Microbiology，2021，87（20）：e0120821.
[38] WOERNER A E， NOVROSKI N M M， WENDT F R， et al. Forensic human identification with targeted microbiome markers using nearest neighbor classification[J]. Forensic Science International Genetics，2019，38：130-139.
[39] TOYOMANE K， YOKOTA R， WATANABE K， et al. Evaluation of CRISPR diversity in the human skin microbiome for personal identification[J]. mSystems，2021，6（1）：e01255-e01320.
[40] KAELBLING L P， LITTMAN M L， MOORE A W. Reinforcement learning： A survey[J]. Journal of Artificial Intelligence Research，1996，4： 237-285.
[41] CHOI R Y， COYNER A S， KALPATHY-CRAMER J， et al. Introduction to machine learning， neural networks， and deep learning[J]. Translational Vision Science and Technology，2020，9（2）：14.
[42] LECUN Y， BENGIO Y， HINTON G. Deep learning [J]. Nature， 2015， 521（7553）： 436-444.
[43] KRIEGESKORTE N， GOLAN T. Neural network models and deep learning[J]. Current Biology，2019，29（7）：R231-R236.
[44] DÍEZ LÓPEZ C， VIDAKI A， RALF A， et al. Novel taxonomy-independent deep learning microbiome approach allows for accurate classification of different forensically relevant human epithelial materials[J]. Forensic Sciencde International Genetics，2019，41：72-82.
[45] DÍEZ LÓPEZ C， MONTIEL GONZÁLEZ D， HAAS C， et al. Microbiome-based body site of origin classification of forensically relevant blood traces[J]. Forensic Science International Genetics，2020，47：102280.
[46] LIU R N， GU Y X， SHEN M W， et al. Predicting postmortem interval based on microbial community sequences and machine learning algorithms[J]. Environmental Microbiology，2020，22（6）：2273-2291.
[47] 中国法医学晋祠宣言[J]. 法医学杂志，2023，39（5）：431-432.

[1]	MA Wenjing, LIU Fan, ZHAO Liang, LIU Hua, PEI Jingzhe, ZHANG Rui, SHI Wei. Study on the Rapid Imaging Diagnosis of Femoral Neck Fracture by Artificial Intelligence [J]. Chinese Journal of Forensic Sciences, 2024, 0(3): 49-54.
[2]	GUO Hongling, MEI Hongcheng, WANG Ping, HU Can, ZHENG Jili, LI Yajun, ZHU Jun. Metal Stable Isotope Techniques and Its Forensic Application [J]. Chinese Journal of Forensic Sciences, 2023, 0(6): 48-55.
[3]	WANG Xu, ZHANG Mingrui, . Study on the Reliability of Forensic Science / Forensic Appraisal Opinions [J]. Chinese Journal of Forensic Sciences, 2023, 0(3): 1-8.
[4]	GUO Hong, YANG Kai, GENG Puyang, LI Yan, LU Qimeng, LING Rong, JIA Rujing. Design and Analysis of Proficiency Testing in the Field of Forensic Appraisal of Electronic Data—Taking the Functional Appraisal of Electronic Data in 2021 as an Example [J]. Chinese Journal of Forensic Sciences, 2022, 0(1): 38-44.
[5]	LIU Haoyang, YANG Xiulei. The Current Situation and Challenges of Electronic Data Forensic Appraisal in Public Security Organs [J]. Chinese Journal of Forensic Sciences, 2022, 0(1): 54-59.
[6]	LI Zhuo, QU Shengqiu, ZHOU Zhilu, CHEN Jiemin, WANG Meng, WANG Maowen, XIA Wentao. Application of CT Measurement in the Identifications of Injury Degree in Medial Orbital Fracture [J]. Chinese Journal of Forensic Sciences, 2022, 0(1): 79-84.
[7]	ZHAO Yuhan, QIAN Huanggui, SHI Shaopei, SUN Nianfeng. Analysis and Research on the Characteristic Change of Peeling Fingerprint [J]. Chinese Journal of Forensic Sciences, 2021, 0(6): 68-71.
[8]	. Reform and Development of Judicial Appraisal in the New Era [J]. Chinese Journal of Forensic Sciences, 2018, 0(1): 1-5.
[9]	. Analysis of 65 Reassessment Cases of Pelvic Fracture Impairment [J]. , 2014, 0(4): 25-27.
[10]	. Molecular Imprinting Technology and its Application in Forensic Science [J]. , 2013, 0(2): 55-58.
[11]	SHEN Min. Leading Scientific Development of Forensic Science Units by Proficiency Testing [J]. Chinese Journal of Forensic Sciences, 2012, 0(6): 19-23.
[12]	YU Yan-lin. The Application of Gold Nanoparticle in the Development of Latent Fingerprint [J]. Chinese Journal of Forensic Sciences, 2012, 0(6): 49-53.
[13]	SHEN Min. Development Trends and Problems in Forensic Toxicology [J]. Chinese Journal of Forensic Sciences, 2012, 0(5): 66-70.
[14]	XIA Qing, FAN Li-hua, WU Jun. Disability Assessment for Hand Dysfunction Caused by Traffic Accident [J]. Chinese Journal of Forensic Sciences, 2012, 0(5): 79-82.
[15]	JIA Xiao-guang. Analysis of Forensic Document Examination in Private Lending Dispute Cases [J]. Chinese Journal of Forensic Sciences, 2012, 0(2): 59-61.