KR20090090476A - Prediction methods for autism using copy number variation and kits by using thereof - Google Patents

Abstract

A method for predicting autism using copy number variation of genome is provided to predict risk factor of autism generation through copy number variation of genome. A method for predicting autism using a copy number variation (CNV) of genome comprises: a step of observing copy number variation of genome; and a step of confirming the expression variation of CNV. A kit for predicting autism comprises a microarray and optical measurement device. The microarray comprises a gene probe for CNV observation related to autism. The optical measurement device measures the expression variation of specific gene on CNV.

Description

Prediction methods for autism using copy number variation and kits by using groom}

The present invention relates to methods and prediction kits for predicting autism by measuring changes in copy number variation ( CNV ).

Autism is a broad syndrome characterized by severe social deficits, impaired communication, repetitive behavior, and limited interest (Bailey et. al ., J Child Psychol Psychiatry (1996) 37 : 89-126; Rutter, J Autism Dev Disord (2005) 35: 241-257). The disease is therefore called autism spectrum disorder ( ASD ) or pervasive developmental disorder ( PDD ) (Veenstra-VanderWeele and Cook, Mol Psychiatry (2004) 9: 819-832).

These heterogeneous disorders are associated with various risk factors ranging from physiology to biological range. Evidence on the genetic background of autism development is presented (Folstein and Rosen-Sheidley, Nat Rev Genet (2001) 2: 943-955; Veenstra-Vander Weele and Cook, Mol Psychiatry (2004) 9: 819-832; Veenstra-VanderWeele et al ., Annu Rev Genomics Hum Genet (2004) 5: 379-405). The concordance rate of autism in monozygotic twins (below 60-90%) is significantly higher than that of dizygotic twins (below 10%), and the relapse rate in brothers and sisters is about 10 times higher than in normal group. (Veenstra-Vander Weele and Cook, Mol Psychiatry (2004) 9: 819-832; Cohen et al ., J Autism Dev Disord (2005) 35: 103-116). ASD is one of several genetic disorders with obvious cytogenetic abnormalities, sometimes referred to as 'secondary' autism (Konstantareas and Homatidis, J Autism Dev Disord (1999) 29: 275-285; Lauritsen et. al ., J Child Psychol Psychiatry (1999) 40: 335-345), a series of cases of autism indicate that it is important to study idiopathic cases to elucidate the genetic mechanisms specific to the autistic phenotype. By limiting cases to idiopathic, the sex ratio was higher than expected and the incidence in the family was higher. This reflects a relatively large contribution of genetic elements (Folstein and Rosen-Sheidley, Nat Rev Genet (2001) 2: 943-955). One of the most preferred genetic models in idiopathic autism is the multilucus model. This is because children inherit multiple genes from parents of heterogeneous phenotypes (Folstein and Rosen-Sheidley, Nat Rev Genet (2001) 2: 943-955). Different combinations of inherited genes may explain polymorphisms for the severity and signs of autism in siblings, sisters and family members. For these complex genetic patterns, various linkage analyzes suggest an association for more than 15 loci in ASD. Recently, a mutation in SHANK3 , a synaptic scaffolding gene, has been shown to be associated with ASD (Risch et. al ., Am J Hum Genet (1999) 65: 493-507; Durand et al ., Nat Genet (2007) 39: 25-27). However, this genetic and phenotypic heterogeneity makes it difficult to find ASD-related genetic mechanisms.

Advances in whole genome association techniques have led to a shift from locus-specific analysis to genome-wide judgments on reproduction. In particular, microarray based comparative genomic hybridization ( Array- CGH ) has overcome the limitations of traditional cytogenetic techniques such as karyotyping and CGH (Speicher and Carter, Nat Rev Genet ( 2005) 6: 782-792). It has also been widely applied to analyze changes in copy number in various diseases. Through array-CGH, a new concept of genomic polymorphism called copy number variation ( CNV ) has been proposed (Iafrate et al ., Nat Genet (2004) 36: 949-951). CNV is believed to reveal phenotypic heterogeneity or various susceptibilities in complex diseases (Freeman et al. al ., Genome Res (2006) 16: 949-961). Recently, Sebat et al. Reported for the first time a significant association between CNVs and idiopathic ASDs, although mutations were not reproducibly observed (Sebat et al. al ., Science (2007) 316: 445-449.

In the present invention, we report that common-CNVs are highly shared among idiopathic ASD patients, regardless of onset or heredity, and to confirm this possibility, we used a full-genomic copy number analysis to determine ASD cases. In addition, all CNVs in the control group were identified and associated.

International Application No. PCT / SE2004 / 000193 (International Application No. 2004.03.02) relates to the diagnosis of autism, with the amino acid sequence SKITHRIHWESASLL having a molecular weight of 1779 +/- 1 Da, 1865 +/- 1 Da and 2022 +/- 1 Da, respectively. A method of diagnosing autism is disclosed by determining the presence of high concentrations of specific peptides that analyze tissue samples, body fluid samples, and / or plasma samples with respect to the presence of specific peptides with SSKITHRIHWESASLL, and SSKITHRIHWESASLLR.

It is an object of the present invention to provide a method for predicting autism and a gene for predicting autism to be used.

In order to achieve the above object, the present invention comprises the steps of 1) observing genome polymorphism ( CnV ) on the chromosome; And (2) provides autism prediction method comprising the step of identifying the expression change of any one or more CNV selected from the group of CNV-related CNV defined in Table 1.

The present invention provides a microarray comprising a CNV observation gene probe for observing autism related CNV groups defined in Table 1 for observing genomic copy number variation ( CNV ) on chromosomes; And (2) provides an autism prediction kit comprising an optical measuring device for measuring the expression change of a specific gene located on the CNV.

The present invention also provides a diagnostic kit comprising a defensin alpha 5 ( DEFA5 ) gene for predicting autism.

A first aspect of the invention comprises the steps of (1) observing genome polymorphism ( CnV ) on chromosomes; And (2) provides autism prediction method comprising the step of identifying the expression change of any one or more CNV selected from the group of CNV-related CNV defined in Table 1. Specifically, the CNV is characterized in that it is present on chromosome 8p23.1 or chromosome 17p11.2. More specifically, the CNV is characterized in that the CNV-L (dielectric polymorphism loss) on chromosome 8p23.1.

As used herein, the terms "chromosome p" and "chromosome q" refer to the short side of the chromosome in the chromosome of the cell division stage where DNA chromosome ends and two chromosomes are bound by centromere. The short-arm or p-arm, the long side is called the long-arm or q-arm, meaning "p-arm" and "q-arm", respectively. "Chromosome 8p23.1" is located in p cancer of human chromosome 8 and "chromosome 17p11.2" is located in p cancer of human chromosome 17. In more detail, staining the human medium-term chromosome shows a characteristic band pattern, which is called a cytogenic band or a cytoband. Humans have 22 autosomal pairs and 1 pair of sex chromosomes. In the late 1960s, scientists found a characteristic staining pattern on the chromosomes of quinacrine mustard stained plants (Caspersson et al., 1968). This banding technique has been applied to human chromosomes (Caspersson et al., 1971), and this band pattern can be refined and not only can the normal chromosomes be easily identified without confusion, but also major structural modifications can also be characterized. It turns out that there is. Regions and bands were consecutively numbered from the mobilization outward of each chromosome arm. Specific unique band landmarks were selected to divide each cancer into regions. These landmarks belonged entirely to the end of the landmark and were considered the first band of that site. Thus, the first landmark found is the beginning of region 2. This site is divided into two or more depending on the staining pattern of the site. For example, the site 8q2 is divided into four bands, and becomes 8q21, 8q22, 8q23, and 8q24. With increased band resolution using chromosomes in the late period, some of the lower resolution bands of the medium chromosome can be further divided according to the emerging high resolution bands. If an existing band is divided into lower subdivisions, the decimal point is placed after the original band position, followed by the number that allocates the lower band. For example, gray staining, high resolution bands are divided into 8q22 bands and subbands 8q22.1, 8q22.2, and 8q22.3. 8q22.1 is closer to isotope than 8q22.3. If the lower band is further divided, additional numbers are added (Kuo-Ho Yen et al., Bioinformatics 2005 21 (17): 3469-3474).

The term "CNV" as used herein refers to copy number variation ( CNV , hereinafter referred to as CNV) on chromosomes. Specifically, there is a partial difference in the amount of genomes in each individual, so there are individual differences in sensitivity to disease and response to drugs.

The terms “CNV-G and CNV-L” described herein refer to the acquisition (CNV-Gain) and deletion (CNV-Loss) of CNV, respectively.

A second aspect of the present invention provides a microarray comprising a CNV observation gene probe for observing autism related CNV groups defined in Table 1 for observing genomic copy number variation ( CNV ) on chromosomes; And (2) provides an autism prediction kit comprising an optical measuring device for measuring the expression change of a specific gene located on the CNV. Specifically, the CNV observation probe is characterized in that the CNV observation probe on chromosome 8p23.1 or chromosome 17p11.2.

A third aspect of the invention provides a kit comprising a defensin alpha 5 ( DEFA5 ) gene for predicting autism.

Hereinafter, the present invention will be described in more detail with reference to Examples. However, the following examples are merely to illustrate the present invention is not limited to the contents of the present invention.

< Example  1> dielectric polymorphism ( copy number variation ; CNV ) analysis

(1) collection of autism patients and cell samples

Twenty-eight children diagnosed with autism spectrum disorder ( ASD ) (24 male and four female, mean age at diagnosis = 5.6 years old) were recruited from the Asan Medical Center's Developmental Disorder Clinic. They are four certified psychiatrists (SC, BK, JH, HY) based on the Diagnostic and Statistical Manual of Mental Disorders, 4th edition ( DSMIV ) Diagnostic Criteria (AP Association, 1994). Was diagnosed by. All study subjects are idiopathic ASD patients with no karyotypic abnormalities or dysmorphic features. We applied a case-control design to identify CNVs that are particularly common in ASD patients compared to normal groups. Normal controls were blood DNA of 62 Korean adults (45 males and 17 females, mean age = 31 years old) without any abnormal signs and family history of genetic disorders. Sample collection and copy number analysis were performed with the approval of the Asan Medical Center Institutional Review Board (2005-0162).

(2) microarray based-comparative genomic hybridization ( array- CGH ) analysis

We used a large insertion clone array consisting of 4,043 human Bacterial Artificial Chromosomes ( BACs ). This was spaced an average of 1 Mb across the entire genome (MacArray ™ Karyo4000, Macrogen, Seoul, Korea). Detailed information about each clone can be found at http://lib.cuk.ac.kr/micro/CGH/MacArray.htm. Genomic DNA extracted from whole blood was used for Array-CGH. Genomic DNA collected from 30 normal men and 30 normal women was used as gender-matched control DNA. Test and control DNAs were labeled using the Bioprime Labeling Kit (Invitrogen, Carlsbad, Calif.). 700 ng of DpnII treated DNA was denatured in 30 ul of 2.5 × random primer solution and rapidly cooled on ice. Cy3, Cy5 labeling and purification were performed according to the manufacturer's instructions. Purified DNA was precipitated using 130 ug of human Cot-1 DNA (Roche, Mannheim, Germany).

Array-CGH was performed as previously described using MAUI hybridization station (BioMicro Systems, Salt Lake city, UT) (Kim et al. al ., Gastroenterology (2006) 131: 1913-1924. Slides were once at room temperature for 10 minutes with solution 1 (2X SSC, 0.1% SDS), twice for 10 minutes with solution 2 (0.1X SSC, 0.1% SDS) and for 1 minute with solution 3 (0.1X SSC). Rinse in succession three times and then rinse in distilled water for 10 seconds. Finally, the slides were spin-dried at 1000 rpm for 2 minutes.

(3) image scanning and data processing

Arrays were scanned using a GenePix 4100B scanner (Axon Instruments, Sunnyvale, Calif.) And images were processed using GenePix Pro 6.0. Processing and standardization of raw array-CGH data was performed using the web-based array-CGH analysis software ArrayCyGHt (http://genomics.catholic.ac.kr/arrayCGH/) (Kim et al ., Bioinformatics (2005) 21: 2554-2555). The signal strength ratio (Cy3 / Cy5) was calculated and converted to log ₂ scale. Five independent self-to-self hybridizations were performed to establish cutoff values for intensity ratios for copy number changes. The standard deviation ( SD ) of the control was 0.081 on a log ₂ scale. In the present invention, the cutoff value for the copy number change was set to ± 0.25, which is three times the control SD on a log ₂ scale. For a more reliable analysis, dye swap analysis was performed in all cases tested. Dye exchange repeated hybridization with the same sample, but fluorescence levels were exchanged to remove dye bias. In other words, in the first array the test DNA was labeled with Cy3 (green) and the control DNA was labeled with Cy5 (red), whereas in the second array the test DNA was labeled with Cy5 and the control DNA was labeled with Cy3. If the copy number change really occurred, the test / control intensity ratio value should change with the replacement of the dye. This consistent copy number change in dye exchange analysis is considered a true change.

(4) results

Dielectric polymorphism ( copy number variation ; CNV ) Confirm

In order to minimize dye bias, we performed a dye-swap analysis to determine CNVs. 1 shows CNVs detected by dye-exchange array-CGH. Based on predetermined cutoff values, 38 parts of CNVs were found across the entire chromosome (Table 1). On average, 3 CNVs for each ASD case and 1.6 CNVs for each control. All 28 ASD cases showed one or more copy number changes, but seven of the 62 controls did not show any CNVs. Of the 38 CNVs, 14 were found in both cases and controls, 19 in ASD cases and 5 in controls (Table 1). The 34-part CNV revealed in the present invention overlaps with previously reported CNV parts, but the remaining 4 copy number changes, although located close to the known CNV part, but did not overlap (Table 1). Several CNVs have been observed to be reproduced in the present invention. Especially 4q35.2 (34/90, 37.8%), 8p23.1 (27/90, 30%), Xq23 (19/90, 21.1%), 13q21.1 (15/90, 16.7%), 1p21.1 CNVs on (12/90, 13.3%), 19q13.12 (11/90, 12.2%) and 17p11.2 (9/90, 10%) were observed more frequently than other CNVs.

Table 1 General characteristics of the identified genome polymorphisms (CNV)

^a dielectric polymorph Polymorphic dielectric from the database (CNV) ID Number: http://projects.tcag.ca/variation/

* The distribution of CNV-L on 8p23.1 and 17p11.2 differs significantly between ASD and control groups.

G, copy number gain; L, copy number loss

Idiopathic ASD Largely associated with CNVs

We compared the distribution of CNVs between ASD and control groups. In order to distinguish CNVs according to the copy number status, the CNV indicating the copy number acquisition was called CNV-G, and the CNV indicating the copy number loss was referred to as CNV-L. Among the 38 parts of CNVs, the CNV-Ls distributions on 8p23.1 ( p = 0.003) and 17p11.2 ( p = 0.008) were significantly different between case and control (Table 1). CNV-L on 8p23.1 was observed in 12 cases (42.9%) in 28 ASDs, but 8 cases (12.9%) in 62 controls. CNV-L on 17p11.2 was observed in four ASDs, but not in the control. When we identified the strength of the association, CNV-L on 8p23.1 showed an important association with idiopathic autism (odd ratio ( OR ) = 5.1, 95% confidence interval ( confidence interval; CI ) 1.7-14.5). We separated men and women to test the strength of association. The association was strong in men (OR = 7.3, 95% CI 2.0-27.1), but not in women, but since most cases are men (86%), it is difficult to say meaningful conclusions about women. The OR for CNV-L on 17p11.2 could not be calculated because it was not in the control group.

CNV-L on 8p23.1 is a 1.1 Mb sized portion comprising three BAC clones (MG4738, MG5256 and MG5126: 6.65Mb-7.71Mb on chromosome 8). This part contains the DEFENSIN gene cluster ( FIG. 1E ). CNV-L on 17p11.2 is approximately 20 kb in size (MG2510: 21.82 Mb-21.85 Mb on chromosome 17) and there are no known coding genes.

< Example  2> 8 p23 .1 Top CNV Quantitative Assessment Analysis for

(1) fluorescence in situ hybridization ( FISH )

Metaphase spreads were prepared from four autism patients and three controls for FISH analysis. The MG5126 BAC clone (8p23.1) was labeled Texas Red and the MG4398 BAC clone (8q24.21) was labeled with fluorescein isothiocyanate ( FITC ). Labels were hybridized simultaneously with medium spreads. FISH signal counts were calculated with 100 microscopic fields per case and the average number was calculated.

(2) Quantification of DEFENSIN Gene by CE-SSCP

DEFENSIN located at 7Mb on 8p One of the gene groups, defensin alpha 5 ( DEFA5 ) and a specific primer for obtaining a myeloperoxidase ( MPO ) gene as an internal control was designed as follows. DEFA5-F: 5-AAAGGGATCTTGAGAACAAA-3 (SEQ ID NO: 1); DEFA5-R: 5-GAGAGCAGGAGTGGATATGT-3 (SEQ ID NO: 2); MPO-F: 5-CCAGCCCAGAATATCCTTGG-3 (SEQ ID NO: 3); MPO-R: 5-GGTGATGCCTGTGTTGTCG-3 (SEQ ID NO: 4). Using these primers, quantitative genomic DNA PCR ( qPCR ) was performed. PCR was denatured for 2 minutes at 95 ° C., then denatured at 95 ° C. for 30 seconds, annealed at 55 ° C. for 30 seconds, and repeated 30 times at 72 ° C. for 30 seconds. Then, the final elongation was performed at 72 ° C. for 7 minutes. PCR products were applied to the previously described capillary electrophoresis-single strand conformational polymorphism ( CE - SSCP ) (Park et al ., Electrophoresis (2006) 27 : 3836-3845). That is, to determine the peak position, 1 ul of qPCR product was mixed with 13.5 ul of deionized formamide and 0.5 ul of ROX 500 standard (Applied Biosystems, Foster City, CA, USA). After thermal denaturation, CE-SSCP was performed with an ABI Prism 310 Genetic Analyzer (Applied Biosystems, Foster City, CA, USA). 1 × TBE buffer containing 3.5% GeneScan polymer and 10% glycerol was used. Electrophoretic conditions were described previously (Park et. al ., Electrophoresis (2006) 27 : 3836-3845). The peak size of the PCR product was analyzed automatically and the peak position was determined using Gene mapper software (Applied Biosystems, Foster City, CA, USA). DEFA5 Peak ratio of to MPO was calculated for both ASD and control cases.

(3) statistical analysis

Differences in the distribution of copy number changes between cases and controls were confirmed using a Chi-square test or Fisher's exact test. DEFENSIN by CNV status To compare the gene mass, one-way ANOVA with Bonferroni correction was used. P values less than 0.05 were considered to be statistically significant.

(4) results

FISH analysis was performed to determine the replication yield of CNV-L on 8p23.1. As a diploid control for chromosome 8, the MG5126 BAC clone representing 8p23.1 and the MG4398 BAC clone representing 8q24.21 were used. Four groups with CNV-L and three groups without CNV-L were tested on 8p23.1. 2 shows the FISH validity results. The double control position (green) showed double signals in both groups, but the number of FISH signals (red) on 8p23.1 in the experimental group with CNV-L was smaller than that without CNV-L. The mean number of red signals per cell (5.0 ± 1.8) in four groups with CNV-L on phase 8p23.1 was significantly lower than that in three groups without 7.1 (2.1 ± 2.1) without CNV-L (p <0.001).

Also located in this part DEFENSIN The amount of defensin alpha 5 ( DEFA5 ), one of the genes in the family, was determined by qPCR based CE-SSCP analysis. Myeloperoxidase ( MPO ) genes were used as internal controls for qPCR and CE-SSCP. The same case was used for FISH analysis and 16 controls, 14 cases without CNV on 8p23.1 and 2 cases with CNV-G in the same portion were tested. DFEA5 the case with the CNV 8p23.1-L The mean signal intensity ratio of MPO (0.11 ± 0.08) was significantly lower than that of the experimental group without 0.5V ± L (0.53 ± 0.10) or with CNV-G (0.66 ± 0.12) ( Fig. 3A ). 3B shows qPCR based CE-SSCP analysis results.

The present invention relates to a method for predicting autism and a prediction kit using genomic polymorphism, and is a technique for predicting autism risk using DNA. Autism screening method using this method is easy to be adopted as autism screening item of general hospital because it is specially designed with low density. When a newborn is born, a copy number variation ( CNV ) test on the chromosome predicts the risk of autism, which is the basis for proactive care in high-risk children. Therefore, the present invention can be applied to regular checkup items of newborns and children in the future. In Korea, when the technology is used as a basic test item for newborns, more than 500,000 tests occur annually, and about 3 million tests are expected annually when expanded to preschool children.

1 shows the results of genome imbalance in autism spectrum disorder ( ASD ) identified through dye exchange hybridization. Test and control DNAs were labeled with Cy3 and Cy5, respectively, and hybridized simultaneously. The same test and control DNAs were then labeled via dye replacement. Blue in the plot represents the test sample profile (test Cy3 / control Cy5 signal strength) and red indicates control (control Cy3 / test Cy5 signal strength). When both dye-exchange profiles appeared to be consistent with peaks over the cutoff range, and at the same time seen converted peaks of similar intensity at the same location, we determined to be true CNVs (black arrows). If the dye-exchange assay did not yield consistent results (red arrow), it was not calculated by CNV. The final judgment on the copy number status (acquired or lost) of the test sample was based on the blue profile. A. CNV-G on 4q35.2; B , CNV-L on 13q21.1; C , CNV-G on 17p11.2; D , CNV-L on Xq23 is shown. E represents CNV-L on 8p23.1 for three consecutive BAC clones containing the 6.6-7.7Mb moiety at 8p23.1. 11 DEFENSIN Family genes are located in this section. The X axis shows the linear position on each chromosome and the Y axis shows the Cy3 / Cy5 signal intensity ratio on a log ₂ scale. G, acquisition; L, lost.

2 shows the quantitative validity results of CNV-L on 8p23.1 phase via FISH. Results of dual color FISH analysis using MG5126 BAC clone (red) on 8p23.1 and MG4398 (green) on 8q24.21. A , in the experimental group without CNV-L on 8p23.1 (above), multiple MG5126 signals were seen, but the signals on 8q24.21 were doubled. B , in the experimental group with CNV-L (below), the number of MG5126 signals was lower than in the experimental group without CNV-L, but the MG4398 signal was consistently doubled.

3 shows the quantitative validity results of CNP status of 8p23.1 phase via qPCR based CE-SSCP. Internal control gene MPO And DEFA5 The quantitative PCR targeting the gene was performed in the experimental group with 8p23.1 CNV-L (n = 4), the experimental station without 8p23.1 CNV (n = 14) and the experimental group with 8p23.1 CNV-G (n = In 2). Signal strength of qPCR product was measured by CE-SSCP. A, DFEA5 in the experimental group with a 8p23.1 CNV-L The average signal intensity ratio of MPO to MPO was lower than that of the test group without 8p23.1 CNV or the test group with CNV-G ( p = 0.000). CNV state on X axis, 8p23.1; Y-axis, average signal strength ratio ( DEFA5 / MPO ). B , CE- for qPCR products from experimental groups with 8p23.1 CNV-L (top plot), CNVs without 8p23.1 CNV (middle plot) and experimental groups with 8p23.1 CNV-G (bottom plot) SSCP chromatogram. Left column, CE-SSCP chromatogram for MPO ; Right column, CE-SSCP chromatogram for DEFA5 .

<110> CATHOLIC UNIVERSITY INDUSTRY ACADEMY COOPERATION FOUNDATION <120> Prediction methods for autism using copy number variation and          kits by using pretty <160> 4 <170> KopatentIn 1.71 <210> 1 <211> 20 <212> DNA <213> Artificial Sequence <220> <223> primer <400> 1 aaagggatct tgagaacaaa 20 <210> 2 <211> 20 <212> DNA <213> Artificial Sequence <220> <223> primer <400> 2 gagagcagga gtggatatgt 20 <210> 3 <211> 20 <212> DNA <213> Artificial Sequence <220> <223> primer <400> 3 ccagcccaga atatccttgg 20 <210> 4 <211> 19 <212> DNA <213> Artificial Sequence <220> <223> primer <400> 4 ggtgatgcct gtgttgtcg 19  

Claims (6)

(1) observing genome polymorphism ( CnV ) on chromosomes; And (2) Autism prediction method comprising the step of identifying the expression change of any one or more CNV selected from the group of CNV-related CNV defined in Table 1. The method of claim 1, Said CNV is present on chromosome 8p23.1 or chromosome 17p11.2. The method of claim 2, Said CNV is CNV-L (dielectric polymorphism loss) on chromosome 8p23.1. (1) a micro array comprising CNV observing gene probes for observing autism related CNV groups defined in Table 1 for observing genomic copy number variation ( CNV ) on chromosomes; And (2) an autism prediction kit comprising a; optical measuring device for measuring the expression change of a specific gene located on the CNV. The method of claim 4, wherein The CNV observation probe is autism prediction kit, characterized in that the CNV observation probe on chromosome 8p23.1 or chromosome 17p11.2. Diagnostic kit containing defensin alpha 5 ( DEFA5 ) gene for predicting autism.

Images