Method of detecting fetal chromosomal aneuploidy

1. A method of detecting chromosomal aneuploidy of a target fetal chromosome in a test sample using a massively parallel sequencing system, the method comprising:

isolating a plurality of nucleic acid fragments from a biological sample of a pregnant woman including fetal nucleic acid fragments as the test sample;

obtaining sequence information (reads) of the plurality of nucleic acid fragments by performing massively parallel sequencing of the plurality of nucleic acid fragments using a massively parallel sequencing system;

mapping the reads obtained using the massively parallel sequencing system to a reference human genome to assign the reads of the plurality of the nucleic acid fragments to the target fetal chromosome;

calculating a GC content and a fraction of reads (Rf) of the plurality of the nucleic acid fragments on the target fetal chromosome to the number of the nucleic acid fragments, based on the reads of the plurality of the nucleic acid fragments assigned to the chromosome;

selecting adaptive reference samples belonging to a shared range of unit values of Rf and unit values of GC content from reference samples, based on the calculated Rf and GC content on the target chromosome;

calculating z scores of verification reference samples and a z score of the test sample using the selected adaptive reference samples;

determining that the target chromosome has chromosomal aneuploidy when, by comparing the calculated z scores of the verification reference samples with the z score of the test sample, the z score of the test sample is larger than the z scores of the verification reference samples; and

performing an invasive method selected from chorionic villus sampling, amniocentesis, and sampling from an umbilical cord, based on the determining that the target chromosome has chromosomal aneuploidy,

wherein the selected adaptive reference samples have a lower coefficient of variance than reference samples without adaptive selection.

2. The method of claim 1 , wherein the biological sample is blood, plasma, serum, urine, saliva, mucus, sputum, feces, tears, or a combination thereof.

3. The method of claim 1 , further comprising excluding intervals with a low confidence level for reads from subjects of analysis by examining depth distribution of the reads of the nucleic acid fragments assigned to the chromosome at each interval, after assigning the reads of the nucleic acid fragments to the chromosome.

4. The method of claim 3 , wherein the interval is an interval set in units of about 5 kb to about 50 kb.

5. The method of claim 3 , wherein the excluding of intervals with a low confidence level for the reads from subjects of analysis comprises removing mismatches, removing multi-mapped reads, removing duplicated reads, or a combination thereof.

6. The method of claim 1 , further comprising performing locally weighted scatterplot smoothing (LOWESS or LOESS) regression analysis of the reads of the nucleic acid fragments according to the following Equation 1 to reduce GC content bias, after assigning the reads of the nucleic acid fragments to the chromosome:

Rf ij′ =RC ij /Σ j=1 22 RC ij   (Equation 1)

wherein Rfij′ represents a corrected fraction of reads on chromosome j in sample i, and RCij represents a corrected number of unique reads on chromosome j in sample i.

7. The method of claim 1 , further comprising performing normalization of the reads of the nucleic acid fragments according to the following Equation 2, after assigning the reads of the nucleic acid fragments to the chromosome:

Rf i′j′ =Rf ij′ /Σ i=1 N Rf ij′   (Equation 2)

wherein Rfi′j′ represents a normalized fraction of reads on chromosome j in sample i, and N represents the total number of samples.

8. The method of claim 1 , wherein the reference sample are obtained from a biological sample of a pregnant woman carrying a euploid fetus.

9. The method of claim 1 , further comprising establishing a linear regression model from all of the reference samples.

10. The method of claim 1 , further comprising extending the unit values of Rf of the reference samples according to Rf values of the test samples, extending the unit values of GC of the reference samples according to GC contents of the test samples, or a combination thereof.

11. The method of claim 1 , wherein the calculating of z scores of the verification reference samples and z scores of the test samples comprises performing a linear regression analysis according to the following Equation 3 and calculating a linear predicted value of Rf according to the following Equation 4:

Rf i′j′ =α+β×GC i′j′ +e   (Equation 3)

Wherein, in Equation 3, Rfi′j′ represents a normalized fraction of reads on chromosome j in sample i, α represents a constant, β represents a coefficient factor between GC content and Rf, and e represents a residual (R); and

Rf′ i′j′ =α+β×GC i′j′   (Equation 4)

in Equation 4, Rfi′j′ represents a fitted predicted value of a fraction of reads on chromosome j in sample i, α represents a constant, and β represents a coefficient factor between GC content and Rf.

12. The method of claim 11 , wherein the calculating of z scores of the verification reference samples and z scores of the test samples comprises calculating a residual (R) from a calculated value from the linear regression analysis and the calculated linear predicted value according to the following Equation 5, and calculating a Z score from the calculated residual according to the following Equation 6:

R=Rf i′j′ −Rf′ i′j′   (Equation 5); and

z score=( R−R ′)/σ′  (Equation 6)

wherein, in Equation 6, R′ represents a mean value of a residual of an adaptive reference sample, R represents a residual value of a test sample, and a′ represents a standard deviation of the residual of the adaptive reference sample.

13. The method of claim 1 , further comprising

selecting reference samples belonging to GC content±unit value of the target chromosome or GC content±unit value of the adaptive reference samples, as verification samples;

calculating z scores of the verification samples; and

verifying that the target chromosome has chromosomal aneuploidy by comparing the calculated z scores of the verification samples with the z scores of the test samples.

14. The method of claim 1 , wherein the target chromosome is chromosome 13, chromosome 18, chromosome 21, an X chromosome, a Y chromosome, or a combination thereof.

15. The method of claim 1 , wherein the chromosomal aneuploidy is trisomy 13, trisomy 18, trisomy 21, XO, XXX, XXY, XYY, or a combination thereof.

16. A system containing a computer-readable medium having recorded thereon a program that performs the steps of:

mapping reads obtained from massively parallel sequencing of a plurality of nucleic acid fragments isolated from a biological sample of a pregnant woman including fetal nucleic acid fragments as a test sample to a reference human genome to assign the reads of the plurality of the nucleic acid fragments to a target chromosome, wherein the massively parallel sequencing was performed using a massively parallel sequencing system;

calculating a GC content and a fraction of reads (Rf) of the plurality of the nucleic acid fragments on the target chromosome to the number of the nucleic acid fragments, based on the reads of the plurality of the nucleic acid fragments assigned to the chromosome;

selecting adaptive reference samples belonging to a shared range of unit values of Rf and unit values of GC content from reference samples, based on the calculated Rf and GC content on the target chromosome;

calculating z scores of verification reference samples and a z score of the test sample using the selected adaptive reference samples; and

determining that the target chromosome has chromosomal aneuploidy when, by comparing the calculated z scores of the verification reference samples with the z score of the test sample, the z score of the test sample is larger than the z scores of the verification reference samples;

wherein the selected adaptive reference samples have a lower coefficient of variance than

reference samples without adaptive selection; and

wherein the result obtained by determining that the target chromosome has chromosomal aneuploidy is used to avoid performing an invasive method selected from chorionic villus sampling, amniocentesis, and sampling from an umbilical cord, based on the determining that the target chromosome has chromosomal aneuploidy.