Postmortem interval deduction method based on microbiome sequencing data and machine learning algorithms

Abstract

The invention discloses a postmortem interval deduction method based on microbiome sequencing data and machine learning algorithms. The method comprises the following steps: by taking DNA (deoxyribonucleic acid) which is extracted from cecum samples of dead bodies of mice as a template, designing primers for 16S rDNA of microorganisms, performing (polymerase chain reaction) amplification constructing a gene library by using amplification products, and performing sequencing; performing OTUs (operational taxonomic units) clustering and species taxonomy analysis on effective sequencing data; andon the basis of abundance distribution situations of species, removing species classes with small relative abundance and interclass variance variation, performing extraction so as to obtain characteristic microorganism varieties for constructing a postmortem interval deduction model, and training a artificial neural network by using sample microorganism abundance data at different time points after death of the mice, so as to obtain a mathematic model which is capable of accurately predicting postmortem intervals.

Description

technical field [0001] The invention belongs to the field of medical examination, in particular to a method for estimating the interval time after death. Background technique [0002] The postmortem interval (PMI) usually refers to the time span from the time of death to the time of autopsy, and it can also be called the postmortem interval or postmortem time. Accurate inference of PMI can assist investigators in determining the direction of investigation, confirming or excluding suspects, estimating the time of crime and reconstructing the crime scene. In addition, PMI inference also plays an extremely important role in certain civil disputes such as insurance claims and property disputes. [0003] The microbial structure is simple, and the growth and reproduction are faster; at the same time, a large part of the microbial community is carried by the body before death, so the initial population is less disturbed by external factors. With the continuous development and pro...

AI Technical Summary

Problems solved by technology

However, it still adopts the currently popular model of first analyzing the relevant microorganisms, and then establishing the relationship between specific microorganisms and the death interval time. At the same time, the sampling time interval is relatively wide, which reduces the credibility of the prediction results

Secondly, the mathematical model adopted in this patent is a linear regression model, which is difficult to avoid the influence of complex and diverse microbial composition and abundance in dead body tissue on the accuracy and effectiveness of the prediction model, which restricts its practical application

Images