| Currently, source identification of body for tissue has been one of important subjects in forensic biology. However, the criteria of Likelihood Ratio in traditional personal identification was not suitable to the source identification of body for tumor tissue, because the high mutation rate of short tandem repeat (STR) locus in which had been sufficiently documented. Consequently, it would be extremely necessary to draw a completely new strategy for source identification of body for tumor tissue.In this study,69 fresh colorectal cancer samples and 31 fresh gastric cancer samples, which had been diagnosed pathologically, had been genotyped with Identifiler system, including 15 STR loci and 1 Amelogenin locus, and the results was paired comparison with those of homologous normal samples to determine whether mutation generated in tumor samples and the mutation type. Four types of mutation, including additional allele (Aadd), new allele (Anew), loss of heterozygous (LOH) and partial loss of heterozygous (pLOH), were found in this group of gastrointestinal tumor tissue samples, three kind of which, including Aadd, Anew and LOH, ineluctably resulted in genotypic alteration of the STR locus (STRGA) with a recall ratio of 3.69% (95%CI:2.82%-4.73%),0.75%(95%CI:0.39%-1.31%) and 0.81% (95%CI: 0.43%-1.39%), respectively. The total mutation rate of STRGA was 5.25%(95%CI: 4.21%-6.46%). STRGA mutation was detected in 32 of the 100 gastrointestinal tumor tissue samples, which indicated the proportion of tumor samples with STRGA was 32.00%(95%CI:23.02%-42.08%). In 87.5%(95%CI:71.01%-96.49%) of the tumor samples with STRGA, the number of STR locus with STRGA was less than 4. The difference of total STRGA rate or the proportion of tumor samples with STRGA between colorectal cancer and gastric cancer samples was not statistically significant (P>0.05).Based on the analysis of the distribution of the number of matched STR locus without identical allele (A0), with 1 identical allele (A1), or with 2 identical alleles (A2) and the number of total identical alleles (IAn) in 2 003 unrelated individual pairs (UIP),280 full sibling pairs (FSP) and the prior genotyped 50 colorectal cancer-homologous normal sample pairs (CRNP),2 groups of Fisher discriminant functions distinguishing CRNP from UIP and 2 groups of Fisher discriminant functions distinguishing CRNP from FSP were established and validated with an error count rate as low as 0.00%(95%CI:0.00%-7.11%) for personal identification with colorectal cancer tissue. Whereafter, all of the 100 pairs of tumor and homologous normal samples were examined with the 4 groups of Fisher discriminant functions and the results indicated that the 4 tumor samples with more than 3 genotypic alteration STR loci could not be assessed source body accurately with an error count rate as low as 4.00%(95%CI:1.10%-9.93%). Based on the above results, a set of discrimination rules was given to source identification for gastrointestinal tumor tissue according to the STR genotype with Identifiler system, which was listed as below:i) Confirmation of surgery or biopsy history of gastrointestinal tumor;ii) Pathological diagnosis for the examined tumor sample;iii) Acquired information of the patient's lineal relatives, especially information about his homozygotic twins or siblings. If necessary, samples of the patient's siblings should be bring into the identification procedure.iv) Satisfied all the above 3 requirements, if IAn=26 and A2=11, the tumor samples can not be excluded coming from the paired comparison normal sample's homologous body; if it was confirmed that the patient had no sibling, the same conclusion can be obtained with the precondition of IAn=23 and A2=8.In order to provide a more precise tool for source identification of body for tumor tissues with more than 3 STRGA loci, insertion/deletion polymorphism (InDel), a kind of biallelic genetic marker, was examined in the above tumor samples. At first, 837 InDel loci was selected from dbSNP with Genome browser of Galaxy system according to a given filter and imported into a sub-database. Then,40 candidate InDel loci was picked out based on a given criteria and a 41plex multiple PCR system, including the 40 candidate InDel loci and 1 Amelogenin locus, was developed and validated.108 unrelated Chinese Han individuals were genotyped with the 41plex multiple PCR system and the allele frequency was obtained.35 of the 40 candidate InDel loci were eventually adopted in the subsequent studies according to the criteria of the minor allele frequency (MAF)=0.250. Results of Hardy-Weinberg Equilibrium (HWE) testing suggested that the allele frequencies distribution of the 35 InDel loci in the group of unrelated Chenes Han individuals had meet the genetic balance. Linkage analysis indicated that those InDel loci on the same chromosome were unlinked with each other. Forensic parameters of the 35 InDel loci were calculated respectively. Range of the MAF, Expected Heterozygosity (He), Polymorphism Information Content (PIC) and Probability of Discrimination Power (PD) were 0.273-0.495, 0.397-0.500,0.318-0.375 and 0.535-0.650, respectively. Cumulated Probability of Discrimination Power (CPD) of the 35 InDel loci was as high as 0.999999987, which was similar with that of Identifiler system.The 100 gastrointestinal tumor samples were genotyped with the 36plex multiple PCR system and paired comparison with that of the homologous normal tissue samples. Consequently, two types of InDel mutation, including pLOH and LOH, were founded in this group of tumor samples. The total recall ratio of InDel genotypic alteration (IDGA) was 0.25%(95%CI:0.11%-0.47%), which was about 1/21 of that of STRGA.At the level of individual, IDGA could be detected in 7.00%(95%CI: 2.86%-13.89%) of tumor samples, which was about 1/4.57 of that of STRGA. Difference of the total recall ratio or the detectable proportion of the tumor samples between IDGA and STRGA was statistically significant with P values of 2.771×10-33 and 1.056×10-5, respectively. Results of the consistency checking suggested that the mutation of the two different genetic markers, InDel and STR, in gastrointestinal tumor tissue was unrelated (P= 1.0000).In conclusion, mutation of STR in gastrointestinal tumor tissue is a common phenomenon. Fisher discriminant functions based on the number of identical alleles or matched STR loci could be a feasible method for source identification of body for gastrointestinal tumor samples. However, the precondition of utilizing those functions is that the number of STR locus with STRGA in the examined tumor sample is not more than 3. The CPD of the developed and validated 36plex multiple PCR system, including 35 InDel loci and 1 Amelogenin locus, is similar with that of Identifiler system. The genetic stability of the 35 InDel loci is significantly higher than STR loci of the Identifiler system in gastrointestinal tumor. Satisfactory complementarity can be shown when the two different genetic markers are simultaneously applied in source identification of body for gastrointestinal tumor samples. |
PDF Full Text Request