The Study On Application Of Combined Genetic Testing Technologies For Laboratory Diagnosis Of Rare Genetic Disease

ObjectivesA rare disease is any disease that affects only a small portion of the population,with an average prevalence 1/500,000,currently considered to be composing about6000 different types,most of which have been found to be a monogenic disorder.Gene testing is one of the most effective methods to diagnose rare genetic diseases.A variety of genetic diseases such as high throughput sequencing,multiplex ligation dependent probe amplification(MLPA),Sanger sequencing,capillary electrophoresis,chromosomal microarray and karyotype analysis have been established in our laboratory.Through technique evaluation,statistical analysis and data mining on the testing results conducted by our laboratory from January 2008(when we started establishing molecular tests for the rare diseases)to October 2016,this study attempts to investigate the detection efficiencies of different assays on gene mutations and the role of combined molecular techniques in the laboratory diagnosis of rare genetic diseases..Materials and MethodsThis study involved with a total of 32,676 cases of rare hereditary diseases,from1721 hospitals in 29 provinces in China,among which 80% of the samples from the Departments of Neurology,Pediatrics,obstetrics and gynecology;18841 were males and 12289 were females,the gender for the rest samples were unclear;the main age group of the patients was young or below(36.2% for infants and 40.1% for adolescents)and a few for middle-aged or the elders(17.5%).The medical exome sequencing based on high-throughput sequencing technology was selected as core assay in this study,covering the exon regions and key splicing sites of 4,813 genes with known functional,and is capable for detecting most point mutations and 80% of the copy number variations.Meanwhile,MLPA,Sanger sequencing,long distance PCR,and methylation specific PCR were used to supplement or confirm the high-throughput sequencing technology.Therefore,our stretegy covers the known genetic basis for most of rare genetic diseases and reveals the relevance with the pathogenic variants.ResultsCausativd variations were detected in 12,064 patients(37.5%)by using variousgene diagnosis techniques,among which the most common pathogenic genes(or loci)were DMD(2072 cases,17.2%),SMN1(1109 cases,9.2%),SCA3(645 cases,5.3%),ATP7B(630 cases,accounting for 5.2%),CYP21A2(570 cases,4.7%),chr15q11q13region(497 cases,4.1%),AR(388 cases,3.2%),NOTCH3(306 cases,2.5%),HTT(266 cases,accounted for 2.2%),DMPK(251 cases,accounted for 2.1%)and so on(the above enumerated only 10 kinds).Overall,for the cases with more specific clinical preliminary diagnosis,the test yield was higher(e.g.,the detection rate for 21 hydroxylase deficiency patients was 56.8%,the detection rate of Wilson disease was57.5%),but the detection rate for complex disease was lower(e.g.,the detection rate for patients with a “epilepsy” diagnosis was 17.5%,The detection rate of the patients with hereditary spastic paraplegia was 26.1%.By reviewing the data of a large number of samples,the frequency of the variants involved in this study is calculated and compared with the frequency of the public database,which can be used to identify some “hot spot” variants in Chinese.NOTCH3 gene data mining is done as a specific case to illustrate this.By reviewing through all the clinical information on the patients,753 samples with a CADASIL diagnosis(CADASIL sample set),3,517 patients with any diagnosis but CADASIL(non-CADASIL sample set)were involved in the study.In addition,24 patients with cerebrovascular disease were diagnosed as a control group of cerebrovascular disease.Then,the author focused on the difference of all the NOTCH3 variations in the CADASIL group and the non-CADASIL group,and,after filtering out the known benign mutation sites,there were remaining 57 suspicious variants(including the previously considered "pathogenic" variants,and the previously uncertain variants),by comparing the frequencies of different sample set and calculating the OR value,it was found that 53 of these mutations appeared only in the CADASIL group,and these variations obtained strong support for pathological variations,3 variants(Val237Met,Aeg75 Gln and Ala1020Pro)have similar frequency in the CADASIL and nonCADASIL patients,thus the 3 variants got strong evidence to be likely benign variants,whereas in published literature these 3 variants were treated as pathogenic variants.The variant(ARG544CYS)is the only one that has been considered pathological in this study,while the frequency in the non-CADASIL group is not zero.The comparative analysis showed that the frequency of Arg544 Cys in CADASIL patients was 0.073,and the frequency in the nonCADASIL patients was 0.0023,and the OR value was 21.989 strongly support the pathogenicity of this variant.This part of the study reflects the importance of large sample size in the detection of rare diseases.ConclusionsThe combined application of NGS and a variety of classical gene test techniques would be instrumental in detecting more genetic-related mutations,which was demonstrated by achieving a total testing yeild of 37.5% in this study,a result higher than the published data obtained by use of the single molecular assays to detect the gene disorders for rare genetic disease.In this study,the frequency of specific gene population was estimated by culling the samples from patients with specific gene-related diseases,and the results showed that this strategy could achieve considerable effect.By applying this strategy,the clinical laboratory for gene diagnostic can benefit more from the accumulation of samples rather than entirely relying on the common mutation database.