Methods and processes for non-invasive assessment of genetic variations

• US 10,699,800 B2
• Filed: 05/23/2014
• Issued: 06/30/2020
• Est. Priority Date: 05/24/2013
• Status: Active Grant

• Current Assignee: Sequenom Incorporated (Laboratory Corporation Of America Holdings)
• Original Assignee: Sequenom Incorporated (Laboratory Corporation Of America Holdings)
• Inventors: Zhao, Chen, Dzakula, Zeljko, Deciu, Cosmin, Kim, Sung Kyun, Mazloom, Amin R., Hannum, Gregory, Ehrich, Mathias
• Primary Examiner(s): Dunston, Jennifer
• Application Number:
  US14/892,782
• Publication Number:
  US 20160224724A1
• Time in Patent Office: 2,230 Days
• CPC Class Codes: C12Q 1/6869 Methods for sequencing
• CPC Class Codes: C12Q 1/6883 for diseases caused by alterations of genetic material
• CPC Class Codes: C12Q 2535/122 Massive parallel sequencing
• CPC Class Codes: C12Q 2537/16 Assays for determining copy number or wherein the copy number is of special importance
• CPC Class Codes: G16B 20/10 Ploidy or copy number detection
• CPC Class Codes: G16B 25/10 Gene or protein expression profiling; Expression-ratio estimation or normalisation
• CPC Class Codes: G16B 30/00 ICT specially adapted for sequence analysis involving nucleotides or amino acids
• CPC Class Codes: G16B 30/10 Sequence alignment; Homology search
• CPC Class Codes: G16B 40/00 ICT specially adapted for biostatistics; ICT specially adapted for bioinformatics-related machine learning or data mining, e.g. knowledge discovery or pattern finding

Abstract

Provided herein are methods, processes and apparatuses for non-invasive assessment of genetic variations that make use of decision analyses. The decision analyses sometimes include segmentation analyses and/or odds ratio analyses.

Claims

1. A computer-implemented method for determining the presence or absence of a microduplication or microdeletion in a fetus, comprising:

(a) receiving input information comprising nucleic acid sequence reads obtained by non-targeted multiplexed massively parallel sequencing of circulating cell-free nucleic acid from a test sample from a pregnant female bearing a fetus, and mapping the nucleic acid sequence reads to portions of a reference genome;

(b) normalizing, using a microprocessor, counts of the nucleic acid sequence reads mapped to portions of the reference genome, wherein the normalizing comprises normalization of guanine and cytosine (GC) bias and an adjustment according to a principal component analysis, thereby providing normalized counts;

(c) segmenting, using a microprocessor, the normalized counts of the portions or the normalized counts in a subset of the portions, thereby providing one or more discrete segments;

(d) identifying, using a microprocessor, a candidate segment among the one or more discrete segments, wherein the candidate segment is identified according to an area under a curve (AUC) analysis, wherein;

i) the AUC analysis is based on a number of portions covered by the segment and an absolute value of a level of normalized counts for the segment, wherein the level corresponds to a negative value of normalized counts for a deletion or a positive value of normalized counts for a duplication; and

ii) the candidate segment has the largest AUC of all the segments on the same chromosome; and

(e) determining the presence or absence of the microduplication or microdeletion according to the candidate segment.

2. The method of claim 1 wherein the normalizing in (b) comprises locally weighted polynomial regression (LOESS) normalization of guanine and cytosine (GC) bias (GC-LOESS normalization).

3. The method of claim 1, comprising filtering one or more portions of the reference genome from consideration and removing the counts in the one or more filtered portions.

4. The method of claim 1, wherein the segmenting comprises circular binary segmentation.

5. The method of claim 4, comprising merging adjacent segments into a single segment.

6. The method of claim 1, comprising determining a log odds ratio (LOR) for the test sample according to a probability of having or not having the microduplication or microdeletion.

7. The method of claim 6, wherein the LOR is the log of the quotient of (i) a first multiplication product of (1) a conditional probability of having the microduplication or a microdeletion and (2) a prior probability of having the microduplication or the microdeletion, and (ii) a second multiplication product of (1) a conditional probability of not having the microduplication or the microdeletion and (2) a prior probability of not having the microduplication or the microdeletion.

8. The method of claim 7, comprising generating a quantification of the candidate segment.

9. The method of claim 8, wherein the quantification comprises a count representation for the candidate segment, a chromosome representation of the chromosome in which the candidate segment is located, or a count representation for the candidate segment and a chromosome representation of the chromosome in which the candidate segment is located.

10. The method of claim 9, wherein:

the quantification is a z-score quantification of the count representation, the chromosome representation, or the count representation and the chromosome representation; and

the z-score optionally is a subtraction product of a (i) test sample count representation less a (ii) median of a euploid count representation divided by a (iii) median absolute deviation (MAD) of the euploid count representation, for the candidate segment, wherein;

the (i) test sample count representation is a ratio of total counts divided by total autosome counts for the test sample, and the (ii) euploid median count representation is the median of a ratio of total counts divided by total autosome counts for euploid samples;

or

the z-score optionally is a subtraction product of a (i) test sample count representation less a (ii) a median of a euploid count representation divided by a (iii) MAD of the euploid count representation, for the chromosome wherein;

the (i) test sample count representation is a ratio of total counts in the chromosome in which the candidate segment is located divided by total autosome counts for the test sample, and the (ii) median of the euploid count representation is the median of a ratio of total counts in the chromosome in which the candidate segment is located divided by total autosome counts for euploid samples.

11. The method of claim 10, wherein:

the conditional probability of having the microduplication or the microdeletion is determined according to fetal fraction determined for the test sample, a z-score of the count representation for the segment determined for the test sample, and a distribution for the fetal fraction of z-scores for the count representation for the segment;

the conditional probability of having the microduplication or the microdeletion optionally is determined by the relationship in equation 23;

12. The method of claim 10, wherein the conditional probability of not having the microduplication or the microdeletion is an intersection between the z-score of the count representation for the segment determined for the test sample and a distribution of z-scores for the count representation for the segment in euploids.

13. The method of claim 10, comprising:

determining the presence of a microdeletion or microduplication if, for the test sample, (i) the z-score quantification of the count representation for the candidate segment, is greater than or equal to the value of 3.95, and (ii) the LOR is greater than zero, and

determining the absence of a microdeletion or microduplication if, for the test sample, (i) the z-score quantification of the count representation for the candidate segment is less than the value of 3.95, and/or (ii) the LOR is less than zero.

14. The method of claim 7, wherein the prior probability of having the microduplication or the microdeletion and the prior probability of not having the microduplication or the microdeletion are determined from multiple samples that do not include the test subject.

15. The method of claim 1, comprising, prior to (a), sequencing the circulating cell-free nucleic acid in the test sample by a non-targeted multiplexed massively parallel sequencing process thereby generating the nucleic acid sequence reads.

16. The method of claim 15, wherein the non-targeted multiplexed massively parallel sequencing process is performed with 1-fold coverage or fraction thereof.

17. The method of claim 15, wherein millions of nucleic acid fragments are sequenced.

18. The method of claim 1, wherein the sequence reads in (a) were generated by a non-targeted multiplexed massively parallel sequencing process performed with 1-fold coverage or fraction thereof.

19. The method of claim 1, wherein the input information in (a) comprises millions of sequence reads.

20. The method of claim 1, wherein the portions of the reference genome to which the nucleic acid sequence reads are mapped are approximately the same length of genomic sequence.