Research On The Strategy To Identify Peripheral Blood From Menstrual Blood Using MiRNA

Objective:Blood stains are the most common body fluid samples in criminal scene.It's important to distinguish blood samples from non-blood samples and to further identify whether they are peripheral blood samples or menstrual blood samples in forensic practice.At present,there is no efficient way to distinguish the two different blood samples.The aim of this study is to construct a mathematical model that can distinguish blood samples from non-blood samples and distinguish peripheral blood samples from menstrual blood samples using the expression difference of multiple microRNAs in two kinds of blood samples(peripheral blood and menstrual blood)and three non-blood samples(saliva,semen and vaginal secretion).Methods:1.In this study,we selected miR-451 a which expressed differently between blood and non-blood samples and three miRNAs(miR-205-5p,miR-214-3p,miR-203-3p)which expressed differently between peripheral blood and menstrual blood samples from previous papers.And we collected 200 five body fluids(peripheral blood,menstrual blood,saliva,semen,vaginal secretion)as training set and verification set.We used SYBR Green qPCR to detect their expression level including RNU6 b,and used SPSS22.0 software to analyze the difference of the ?Ct values of the training set and to establish two Fisher discriminant functions.The “blood samples” and “non-blood samples” were used as dependent variables and the ?CT values of miR-451 a were used as independent variables in the first discriminant function.The “peripheral blood” and“menstrual blood” were used as dependent variables and the ?CT values of three miRNAs(miR-205-5p,miR-214-3p,miR-203-3p)were used as independent variables in the second discriminant function.Two Fisher discriminant functions can distinguish blood samples from non-blood samples and distinguish peripheral blood samples from menstrual blood samples gradually.The verification set was used to verify the accuracy of the model.2.On the basis of our previous study,we continued to collect 250 forensic body fluid samples and added another two miRNAs(miR-144-3p,miR-144-5p).“Peripheral blood”,“menstrual blood” and “non-blood body fluids(saliva,semen and vaginal secretion)” were used as dependent variables,and the ?Ct values of six miRNAs(miR-451 a,miR-144-3p,miR-144-5p,miR-205-5p,miR-214-3p,miR-203-3p)in training samples were used as independent variables to build Fisher discriminant analysis model which can identity blood with non-blood samples and also peripheral blood with menstrual blood samples at the same time.The verification set was used to verify the accuracy of the model.3.In order to compare two Fisher discriminant analysis model and verify whether the discriminant analysis model of peripheral blood and menstrual blood based on samples from single source body fluid is suitable for the identification of mixed samples,8 samples of peripheral blood and 8 samples of menstrual blood were collected and used to prepare mixed samples.The expression of six miRNAs(miR-451 a,miR-144-3p,miR-144-5p,miR-205-5p,miR-214-3p,miR-203-3p)were measured.The ?CT values were substituted into two discriminant analysis models to compare the accuracy of the two discriminant analysis models in identifying single and mixed samples of peripheral blood and menstrual blood.Results:The two discriminant analysis models which were used to distinguish blood from non-blood samples and also peripheral blood from menstrual blood samples weresuccessfully constructed.The self-validation accuracy of the model was 100%,the cross-validation accuracy was 100%,and the model validation accuracy was 100% in the first discriminant analysis model(stepwise method).The self-validation accuracy of the model was 99.5%,the cross-validation accuracy was 99.5%,and the model validation accuracy was 100% in the second discriminant analysis model(non-stepwise method).These two discriminant analysis models both had high accuracy in the identification of samples from single source body fluid.In addition,the comparison results of the two discriminant analysis models showed: the samples from single body fluid were all identified correctly by the two models,but the results of identification of mixed sample by the two models were both different from the theoretical results,so these two models were only suitable for the identification of the samples from single body fluid..Conclusion:The two discriminant analysis models of peripheral blood and menstrual blood samples constructed in this study all had high accuracy.People can choose the suitable discriminant model according to the sample size and experimental conditions or use these two methods to identify the body fluid of unknown samples and confirm each other at the same time.This study provides a scientific and accurate identification strategy to distinguish blood samples from non-blood samples and distinguish peripheral blood samples from menstrual blood samples,and has a great potential to be used in forensic practice.