Wound Age Estimation Of Multiple Biomarkers:a Preliminary Study Using Multivariate Analysis

Objective:To construct classification and regression models for estimating wound age,based on the expression of 14 genes(Abhd2,Prr3,Trit1,Arid5 a,Ier3,Rcc1,Rae1,Impact,Tmem45 b,Lin37,Dennd5 a,Fam210a,Myg1 and Lrrc41)which was explored by RT-qPCR.Methods:A total of 78 Sprague–Dawley rats were divided randomly into a control group and contusion groups at 4,8,12,16,20,24,28,32,36,40,44,and 48 h post-injury(n = 6 per group).A counterpoise fell freely through a clear Lucite guide tube onto the right posterior limb causing the skeletal muscle contusion.The expression levels of the target mRNAs were calculated using the statistical model(1 + Eff.)-△△Ct,normalized with the geometric mean of the reference gene(RPL13 and RPL32 mRNAs)levels.Multivariate statistical analyses including partial least squares discrimination(PLS-DA),Fisher discriminant analysis(FDA)and partial least squares regression(PLSR)were employed to establish mathematical models for wound aging.Results:The 14 genes which were involved in wound healing were differentially expressed after skeletal muscle contusion in rat,showing a good correlation of wound age.There were great differences between the groups(control group,4-24 h and 28-48h),which were assessed by partial least squares discriminant analysis(PLS-DA).And the 78 samples were regrouped according to the results of PLS-DA and practice in forensic pathology.The classification results of FDA for control group plus two wound age groups(4–24 and 28–48 h)showed that 97.4% cross-validated grouped case and 76.7% of the ungrouped cases were classified correctly.The classification accuracy of external validation was 83.3% when the wound age was 4– 24 h;the accuracy for the 28–48 h wound age was 58.3% as the all control group cases were all clustered correctly.When the data for control group plus three wound age groups(4-12 h,16-24 h,28-48h)were explored by enter independence together,94.9% cross-validated grouped case were discriminated correctly.The classification accuracy of external validation was similar as that of control group plus two wound age groups(4–24 and 28–48 h),76.7% of the ungrouped cases classified correctly(control group: 100%,4-12h: 83.3%,16-24h: 83.3%,28-48h: 58.3%).The PLS regression prediction model showed reasonable internal predictive validity,which was reflected by its relative higher R2(0.71)and relatively lower root mean squared error(RMSEcv=8h).The prediction results of the PLS model showed lower relatively external predictive validity(RMSEp = 11 h,R2=0.52)than internal predictive validity.From the results of classification and regression model,it seems that FDA prediction model was more suitable for wound aging.Conclusion:The mathematical models,which built based on the expression signatures of 14 selected genes,has great potential to estimate the wound age of skeletal muscle contusions in rats.Moreover,the finding of the study emphasized the importance of the multivariate techniques employed herein and provides new information to develop strategies for future wound age estimation.