Array-based targeted copy number detection

Abstract

Array-based targeted copy number detection, such as detection of contaminated samples and / or variable concentration samples, including obtaining a set of intensity signals from an assay of a set of input samples; performing cross-sample calibration on the intensity signals based on a reference sample, the calibration including constructing a reference signal distribution based on intensity signals of the reference sample and calibrating a set of intensity signals corresponding to one or more input samples based on the reference signal distribution for the input samples; determining, for the one or more input samples and from one or more respective sets of calibration intensity signals corresponding to the one or more input samples, at least one respective aggregated calibration signal from the targeted genomic region to produce an aggregate of aggregated calibration signals; and detecting variants in the targeted genomic region based on the aggregated calibration signal aggregate.

Description

Background Art

[0001] Copy number detection can be performed with various types of arrays for genotyping, cytogenetics or methylation. As an example, the Infinium BeadArray technology provided by Illumina, Inc., San Diego, California supports two modes of copy number detection: discovery mode and targeted mode. For the discovery mode, copy number variation / variant (CNV) events can be detected in unknown regions of the genome in an unbiased manner, while in the targeted mode, copy number changes (i.e., CNV) detection is focused on specific genomic regions of interest.

[0002] CNVs are implicated in many types of human diseases, such as neuropsychiatric disorders, developmental disorders, cardiovascular diseases, autoimmune diseases, and cancer. Therefore, copy number detection assays have been used in clinical applications such as cytogenetics, vector screening, pharmacogenomics, and precision medicine. Copy number detection has also proven useful in veterinary genetics and other non-human genetics applications. Summary of the invention

[0003] By providing a computer-implemented method, the shortcomings of the prior art are overcome and additional advantages are provided. The method includes obtaining an intensity signal aggregate from an assay of an input sample array comprising genetic material, and performing a cross-sample calibration on the intensity signals of the intensity signal aggregate based on one or more reference samples. Performing a cross-sample calibration includes: constructing a reference signal distribution based on the intensity signals of one or more reference samples; and for one or more input samples in the input sample set: obtaining a corresponding intensity signal set corresponding to the input sample in the intensity signal aggregate, the intensity signal set corresponding to the input sample including (i) a first subset C of intensity signals from one or more targeted genomic regions of interest, and (ii) a second subset B of intensity signals from at least one genomic region outside the one or more targeted genomic regions of interest, and calibrating the intensity signals in C based on the reference signal distribution to generate a corresponding calibration intensity signal set corresponding to the input sample. The method also includes determining at least one corresponding aggregate calibration signal from one or more targeted genomic regions of interest for one or more input samples and based on one or more corresponding calibration intensity signal sets corresponding to the one or more input samples, wherein the determination generates an aggregate calibration signal aggregate; and detecting one or more variants in one or more targeted genomic regions of interest based on the aggregate calibration signal aggregate.

[0004] Further, a computer system is provided, the computer system comprising a memory and a processor in communication with the memory, wherein the computer system is configured to perform a method for improved calling of copy number variants in a genomic sequence. The method comprises obtaining an intensity signal aggregate from a determination of an input sample array comprising genetic material, and performing a cross-sample calibration on the intensity signals of the intensity signal aggregate based on one or more reference samples. Performing a cross-sample calibration comprises: constructing a reference signal distribution based on the intensity signals of one or more reference samples; and for one or more input samples in the input sample set: obtaining a corresponding intensity signal set corresponding to the input sample in the intensity signal aggregate, the intensity signal set corresponding to the input sample comprising (i) a first subset C of intensity signals from one or more targeted genomic regions of interest, and (ii) a second subset B of intensity signals from at least one genomic region outside the one or more targeted genomic regions of interest, and calibrating the intensity signals in C based on the reference signal distribution to generate a corresponding calibration intensity signal set corresponding to the input sample. The method also includes determining, for one or more input samples and based on one or more corresponding calibration intensity signal sets corresponding to the one or more input samples, at least one corresponding aggregated calibration signal from one or more targeted genomic regions of interest, wherein an aggregated calibration signal aggregate is determined; and detecting one or more variants in the one or more targeted genomic regions of interest based on the aggregated calibration signal aggregate.

[0005] Still further, a computer program product is provided, the computer-readable storage medium can be read by a processing circuit and store instructions, the instructions are executed by the processing circuit to perform a method for improved calling of copy number variants in a genomic sequence. The method includes obtaining an intensity signal aggregate from a determination of an input sample array containing genetic material, and performing a cross-sample calibration on the intensity signals of the intensity signal aggregate based on one or more reference samples. Performing a cross-sample calibration includes: constructing a reference signal distribution based on the intensity signals of one or more reference samples; and for one or more input samples in the input sample set: obtaining a corresponding intensity signal set corresponding to the input sample in the intensity signal aggregate, the intensity signal set corresponding to the input sample including (i) a first subset C of intensity signals from one or more targeted genomic regions of interest, and (ii) a second subset B of intensity signals from at least one genomic region outside the one or more targeted genomic regions of interest, and calibrating the intensity signals in C based on the reference signal distribution to generate a corresponding calibration intensity signal set corresponding to the input sample. The method also includes determining, for one or more input samples and based on one or more corresponding calibration intensity signal sets corresponding to the one or more input samples, at least one corresponding aggregated calibration signal from one or more targeted genomic regions of interest, wherein an aggregated calibration signal aggregate is determined; and detecting one or more variants in the one or more targeted genomic regions of interest based on the aggregated calibration signal aggregate.

[0006] In one or more embodiments, calibration of the intensity signal in C corresponding to the set of intensity signals of an input sample includes establishing a mapping for the input sample based on a relationship between (i) the intensity signal in B and (ii) a reference signal distribution.

[0007] In one or more embodiments, establishing a mapping includes defining a mapping function M(x) such that M(x) maps the intensity signal x to: for x present in B, M(x) = a matching intensity signal of a vector A of reference signal intensities from a reference signal distribution, the reference signal intensities corresponding to at least one genomic region outside one or more targeted genomic regions of interest; for x not present in B but falling between multiple intensity signals in B, M(x) = a linear interpolation of the M(x) mapping based on the multiple intensity signals in B; and for x not present in B and not falling within the range of intensity signals in B, M(x) = an extrapolation of the mapping based on the highest quantile and the lowest quantile in B.

[0008] In one or more embodiments, constructing a reference signal distribution calculates vector A as the cross-sample median of autosomal array probes outside of one or more targeted genomic regions of interest.

[0009] In one or more embodiments, calibrating the intensity signals in C further comprises mapping the intensity signals in C using a mapping function to generate a set of calibration intensity signals corresponding to the input sample.

[0010] In one or more embodiments, obtaining the intensity signal aggregate includes, for the input sample set, identifying probe hybridization bias using the array hybridization control probe set by aggregating the row-based normalized raw intensity values ​​from the control probes into an aggregate value c s , the row-based normalized intensity values ​​from assays targeting human genomic material are aggregated into an aggregate value x s , and the pollution factor f s Determine x s and c s A function where f s 、x s and c s Determined based on input sample.

[0011] In one or more embodiments, the contamination factor f s The function is: f s =x s / c s .

[0012] In one or more embodiments, determining at least one corresponding aggregation calibration signal for one or more input samples and based on one or more corresponding calibration intensity signal sets corresponding to the one or more input samples includes: for the aggregation calibration signal of at least one aggregation calibration signal, determining a first aggregation signal from a calibration intensity signal set corresponding to a targeted region of the input sample, and using a contamination coefficient to correct the first aggregation signal and generate a second aggregation signal, wherein the second aggregation signal is output as the aggregation calibration signal of the targeted region of the input sample.

[0013] In one or more embodiments, using the contamination coefficient and generating a second aggregate signal includes (i) using a regression-based model to predict the contribution of contamination based on the contamination coefficient, (ii) determining a residual based on the first aggregate signal and the contribution of contamination predicted by the model, and (iii) determining a second aggregate signal based on the residual and a composite contamination coefficient from the input sample.

[0014] In one or more embodiments, the one or more variants are one or more copy number variants.

[0015] In one or more embodiments, prior measurements of (i) deoxyribonucleic acid (DNA) quantification of the input sample, (ii) normalization of the input sample, and (iii) the fraction or amount of DNA contaminants in the input sample are not known or required in practicing the method.

[0016] In one or more embodiments, the input samples of the input sample set contain at least one of (i) varying amounts or concentrations of deoxyribonucleic acid (DNA) relative to each other or (ii) varying fractions of contaminant DNA relative to each other.

[0017] In one or more embodiments, the aggregate of intensity signals is derived from a high throughput genotyping platform that genotypes an input sample using a microarray-based genotyping platform.

[0018] Additionally, a computer-implemented method includes obtaining an aggregate of intensity signals from an assay of a set of input samples comprising genetic material, identifying probe hybridization bias using a set of array hybridization control probes by aggregating row-based normalized raw intensity values ​​from the control probes into an aggregate value c s , the row-based normalized intensity values ​​from assays targeting human genomic material are aggregated into an aggregate value x s , and the pollution factor f s Determine x s and c s A function where f s 、x s and c s The contamination factor is determined based on an input sample of the input sample set, and a first signal obtained based on an intensity signal of the intensity signal set is corrected using the contamination factor to generate a corrected signal.

[0019] In addition, a computer system is provided, comprising a memory and a processor in communication with the memory, wherein the computer system is configured to perform a method comprising: obtaining an aggregate of intensity signals from an assay of a set of input samples comprising genetic material, identifying probe hybridization bias using a set of array hybridization control probes by aggregating row-based normalized raw intensity values ​​from the control probes into an aggregate value c s , the row-based normalized intensity values ​​from assays targeting human genomic material are aggregated into an aggregate value x s , and the pollution factor f s Determine x s and c s A function where f s 、x s and c s The contamination factor is determined based on an input sample of the input sample set, and a first signal obtained based on an intensity signal of the intensity signal set is corrected using the contamination factor to generate a corrected signal.

[0020] Still further, a computer program product is provided, the computer readable storage medium being readable by a processing circuit and storing instructions, the instructions being executed by the processing circuit for performing a method comprising improved calling of copy number variants in a genomic sequence. The method comprises obtaining an aggregate of intensity signals from an assay of a set of input samples comprising genetic material, using a set of array hybridization control probes to identify probe hybridization bias by aggregating row-based normalized raw intensity values ​​from the control probes into an aggregate value c s , the row-based normalized intensity values ​​from assays targeting human genomic material are aggregated into an aggregate value x s , and the pollution factor f s Determine x s and c s A function where f s 、x s and c s The contamination factor is determined based on an input sample of the input sample set, and a first signal obtained based on an intensity signal of the intensity signal set is corrected using the contamination factor to generate a corrected signal.

[0021] In one or more embodiments, using the contamination coefficient and generating a correction signal includes (i) using a regression-based model to predict the contribution of contamination based on the contamination coefficient, (ii) determining a residual based on the first signal and the contribution of contamination predicted by the model, and (iii) determining a correction signal based on the residual and a composite contamination coefficient from an input sample.

[0022] In one or more embodiments, the contamination factor f s The function is: f s =x s / c s .

[0023] In one or more embodiments, the first signal is a first aggregation signal from an ensemble of intensity signals of an ensemble of intensity signals, the first aggregation signal corresponding to a target region of an input sample of the set of input samples, and wherein the corrected signal is a corrected aggregation signal.

[0024] In one or more embodiments, the aggregate of intensity signals is derived from a high throughput genotyping platform that genotypes an input sample using a microarray-based genotyping platform.

[0025] Additional features and advantages are achieved through the concepts described herein. BRIEF DESCRIPTION OF THE DRAWINGS

[0026] The various aspects described herein are particularly pointed out and clearly claimed as examples in the claims at the end of this specification. The foregoing and other objects, features and advantages of the present disclosure will become apparent from the following detailed description taken in conjunction with the accompanying drawings, in which:

[0027] Figure 1 Depicted is an example approach to constructing microarrays for targeted copy number detection using a high-throughput genotyping platform;

[0028] Figures 2 to 3 Example effects of contamination and total target DNA amount on signal intensity are shown;

[0029] Figures 4 to 6 An example comparison between the capabilities of aspects described herein and existing methods for CNV calling is shown;

[0030] Fig. 7A Depicted is an example process for array-based targeted copy number variant detection according to aspects described herein;

[0031] Figure 7B Depicted are example processes for contamination coefficient based signal correction according to aspects described herein; and

[0032] Figure 8 Examples of computer systems and related devices that incorporate and / or use aspects described herein are depicted. DETAILED DESCRIPTION

[0033] Methods for array-based targeted copy number detection are described herein, such as detection of contaminated samples and / or samples of variable concentrations. For example, the exemplary methods can enable, facilitate, and provide accurate copy number determination of samples, including, for example, samples having (i) variable amounts of input deoxyribonucleic acid (DNA) and / or (ii) variable fractions of contaminant DNA.

[0034] There are methods for targeted copy number detection using high throughput genotyping platforms. In one example, a microarray-based genotyping platform is used. Figure 1 Depicted is an example method for constructing a microarray for targeted copy number detection using a high-throughput genotyping platform. Figure 1 , initially a silicon wafer 102 is obtained and at 110, a photoresist 104 is placed / patterned onto a base layer 106 of the wafer. Plasma etching (112) etches microwells into the base 106. This is followed by a cleaning step 114 to remove the photoresist 104. A pattern / array of wells is formed into which beads 108 are placed to create (116) a microarray.

[0035] By way of example, the BeadArray microarray technology provided by Illumina, Inc. of San Diego, California is used. The BeadArray microarray technology uses silica microbeads where hundreds of thousands to millions of genotypes can be determined at once for a single individual on the surface of each array, or BeadChip. The tiny silica beads are housed in carefully etched microwells and coated with multiple copies of oligonucleotide probes that target specific loci in the genome. As the DNA fragments pass through the BeadChip, each probe binds to a complementary sequence in the sample DNA, terminating one base before the locus of interest. Allele specificity is conferred by a single base extension incorporating one of four labeled nucleotides. When excited by a laser, the nucleotide labels emit a signal. The intensity of this signal conveys information about the allele ratio at that locus.

[0036] Detection of genetic variants using contaminated and / or variable concentration saliva DNA samples :As a non-invasive source of DNA, saliva is an important sample type in genomics. Analysis of saliva DNA enables conventional direct-to-consumer (DTC), research and / or clinical genomics applications. However, due to the presence and variability of non-human contaminant DNA and the variability of total DNA concentration, the saliva sample type has some unique challenges for accurate genetic variant detection. For example, it has been shown that the false positive rate of SNV detection is slightly higher in saliva and oral samples, while the sensitivity of CNV detection for saliva samples is 25% lower than that for blood, and it has been shown that in the case of whole genome sequencing, more than 95% of SNVs found in saliva are consistent with paired blood samples, while for CNV, only 75% are consistent. In general, CNV detection is more challenging when processing saliva samples compared to blood samples.

[0037] Aspects described herein propose methods for solving the saliva-specific problem of CNV detection, so that more accurate CNV calls for genetic applications can be achieved, such as pharmacogenomics and vector screening (as an example) for saliva samples as DNA sources. For example, various aspects can achieve more accurate CNV detection based on saliva DNA in the case of unknown / variable DNA concentration and / or unknown / variable contamination fraction. It is advantageously not required to input DNA quantification or normalization of DNA samples, or to pre-measurement of the fraction or amount of DNA contaminants. At the same time, it supports sample collection, each sample having (i) DNA of different amounts / concentrations and / or (ii) contaminant DNA of different fractions. According to aspects presented herein, CNV detection performed using saliva sample collection can work almost as well as detection using non-polluting normalized DNA sample collection.

[0038] Signal aggregation with target genomic regions: To enable CNV detection for a specific target region in the genome, and as a specific example, a variable number of target-specific 50-mer DNA probes are designed to provide complementary assays with 3' DNA ends spanning the target region. Intensity signals from all assays are captured by a scanner (e.g., the iScan system provided by Illumina), and then individual assays are normalized, weighted, and aggregated.

[0039] Specifically, for the target region t of sample s, the aggregation signal x ts Given by equation 1:

[0040]

[0041] Where i indicates the determination, w ti is the assay-specific weight, and x tis is the intensity or normalized intensity signal of a particular assay i in sample s (e.g., R is the total intensity from the red and green channels, or LRR is the log R ratio).

[0042] A. Cross-sample calibration based on the intensity signal of the reference sample :

[0043] Contamination results in differences in intensity distributions between samples with different levels of contaminants and human DNA quality. In a first aspect, a method is provided to correct for systematic signal differences by (i) constructing a reference signal distribution and (ii) performing a reference-based calibration of the sample. The quantile normalization algorithm can operate across a chip containing multiple samples.

[0044] Reference intensity distribution is constructed by identifying reference sample set, wherein low-quality samples have been removed.On the reference set, process calculation reference intensity vector A (vector of reference intensity values) is used as the cross sample median of the autosomal array probe (e.g., Infinium genotyping assay) outside the target region of interest in the sample. Reference intensity value is further sorted to form A as a reference quantile vector, i.e., as the sorting / ranking list of intensity value. Missing value can be retained as the lowest quantile during the construction of this reference quantile vector, and therefore, the signal intensity of the low quantile corresponds to missing value.

[0045] The following process can be used to calibrate new samples. For each new sample, the process divides the array signal of the sample into two subsets: a subset B containing the intensity signals of all autosomal array probes (e.g., Infinium genotyping assays) outside the target region of interest of the sample; and a subset C containing all the remaining intensity signals of the sample (i.e., not included in set B). Then, using A and B, the process defines a mapping M of signal intensity quantiles, i.e., a function for calibrating any given intensity value x in the sample, as follows:

[0046] For any intensity value x in B, M(x) is defined as the matching quantile value in A (since B can be sorted in the same way as the quantile vector is formed);

[0047] For any intensity value x that is not in B but falls within the range of intensity values ​​in B (i.e., between two adjacent values ​​in B), M(x) is defined as a linear interpolation based on the values ​​in B (e.g., a local interpolation of the value in B that is closest to x); and

[0048] For any intensity value x outside the range of intensity values ​​in B, M(x) is defined as an extrapolation based on the highest and lowest quantiles in B (e.g., the first 100 highest and last 100 lowest intensities, as an example).

[0049] Once a map M is defined based on A and B, the process calibrates each intensity value x in C to M(c) to produce calibrated individual intensities that can be used for target copy number detection. Specifically, for each CNV target region t in sample.s, the aggregate signal (x ts ), as substituted into the above equation 1, that is, as shown in the following equation 2:

[0050]

[0051] In equation 2, x in equation 1 is tis The term (“rule strength”) has been replaced by M(x tis Based on this, the copy number can be determined, and the x in equation 2 ts This provides a focused calibration signal for the region t of the sample s.

[0052] Thus, as a first aspect, the cross-sample calibration of Section A provides a cross-sample intensity signal calibration based on a reference sample intensity signal. In embodiments, this aspect can be used in combination with other aspects described herein, such as the aspects described below in Section B (control-based sample-specific contamination adjustment), but such use together is optional. In other words, if desired, each aspect (Section A, Section B) can be used separately / independently / separately from each other.

[0053] B. Sample-specific contamination adjustment based on internal controls :

[0054] Sample-specific contamination adjustment based on controls involves adjustment to account for variable contamination levels in a given sample. This aspect attempts to estimate the contamination level in the sample directly, and then adjusts the aggregation signal (which may optionally be an aggregation calibration signal as discussed in Section A above) based on the estimated contamination level.

[0055] For each sample, an array hybridization control probe set (e.g., Infinium genotyping assay) is used to achieve contamination correction. The total probe hybridization deviation from the experiment itself (e.g., reagents and other assay conditions) rather than from human DNA or non-human DNA contamination is obtained using the control. In other words, the control can independently measure the assay efficiency in the DNA input by measuring the assay itself in addition to the DNA amount in the assay. The control intensity is subjected to a normalization procedure, such as a row normalization procedure according to the aspects presented herein. Specifically, the process normalizes the original intensity by removing some metrics (such as median) of the samples in the same row on the array and then adding back a global metric (e.g., global median). Then, the process aggregates the row normalized intensities of all hybridization probes into an aggregate value c s .

[0056] Using the same general approach as for the control, the process aggregates all assays targeting human autosomes into an aggregate value x s , as a measure for obtaining the overall human DNA abundance. Thus, the x measure provides the intensity from the other (ie, other than the control) probes, and thus x s and c s The ratio of the values ​​is used to represent the proportion of human DNA in the sample. Therefore, the control adjusted contamination (CAC) factor is determined as f s =x s / c s .

[0057] The process can then use the CAC coefficients to adjust the target-specific array signal x in a sample-specific manner. ts , and then assign the copy number. In the embodiment where this contamination adjustment is used in conjunction with the cross-sample calibration method of Section A above, the adjusted x is obtained. ts It can be x in equation 2 ts (Aggregate calibration signal). For example, a cross-sample calibration based on the intensity signal of the reference sample is performed, wherein a reference signal distribution is constructed, intensity signals corresponding to the input samples are obtained and divided into sets B and C, and the intensity signals in C are calibrated as discussed above. Then, after performing this cross-sample calibration, the CAC coefficients can be used to adjust the signals, such as the individual calibration signals and / or the aggregate signal x ts Alternatively, if the control-based sample-specific contamination adjustment of Section B is not used in conjunction with the cross-sample calibration approach of Section A, the adjusted x is ts It can be x in equation 1 ts .

[0058] Use CAC coefficients to adjust target-specific array signals x tsEstimate the impact of contamination on the observed signal. In particular, build a regression-based model, such as a linear or nonlinear machine learning model or any complex prediction model to derive the CAC metric f from all samples. s Predicted target strength signal (x ts ) in order to determine the observed signal (x) of the contamination on the sample ts ). Removing this contribution from the observed signal provides the "residual" (adjusted intensity signal) as the signal observed and attributable to copy number, which can be a major part of the information of interest.

[0059] Therefore, the process updates the target strength signal as the residual of the predictor, using Equation 3 as follows:

[0060] r ts =x ts -Model(f s )(Equation 3)

[0061] Over all samples s, as the residual r of the offset m ts The bias is multiplied by the model (f s ) to obtain the adjusted / corrected intensity signal (x' ts ):

[0062] x′ ts =r ts + median {model (f ＊ )} (Equation 4)

[0063] Then use the corrected intensity signal (x' ts ) to determine the copy number. This is in contrast to using the initial intensity signal on the sample (i.e., x in Equation 1 or Equation 2). ts ).

[0064] Figures 2 to 3 shows example effects of contamination and total target DNA amount on signal intensity, and Figures 4 to 6 An example comparison between the capabilities of the aspects described herein and existing methods for CNV calling is shown. Specifically, Figure 2 A graphical representation of the effect of contamination on the aggregate intensity signal across the target true copy number is depicted. It can be seen that the level of contamination in the saliva sample directly affects the signal range and signal-to-noise ratio (SNR) from the array, and therefore adversely affects the copy number assignment. For illustrative purposes, an artificial saliva sample was used in this analysis, and the specific target shown is the CYP2D6 5' flanking region.

[0065] Figure 3Depicts a graphical representation of the effect of total target DNA input on intensity signal variability and deviation across three runs. Here, Infinium measures (as an example) that intensity signal variation increases as human DNA input decreases. The same set of real saliva samples is run under three different conditions. Under each such condition, a significant correlation is observed between total target DNA amount and signal variability.

[0066] Figure 4 Depicted is a graphical representation of a comparison of CNV calling accuracy (represented by the F-measure) between (i) a CNV detection method as disclosed herein ("Current Method", covering both aspects of Sections A and B above) and (ii) a conventional CNV detection method ("Prior Method") in both (a) a cell line DNA sample (i.e., without contamination) and (b) a human saliva DNA sample (i.e., with contamination). As can be seen, the Current Method provides equal / improved CNV calling accuracy for (a) and significantly improved CNV calling accuracy for (b). It should be noted that in the accuracy evaluations here, the degree of contamination was not relevant, all saliva samples were included, but filtering the saliva samples resulted in further improvements in accuracy levels.

[0067] Figure 5 Two graphical representations 502, 504 are depicted. The first graphical representation (502, at the top) provides a representation of an artificially contaminated sample produced with a mixture of human DNA and varying degrees of contaminants (ranging from 10% to 90%, centered at about 50%). The second graphical representation (504, at the bottom) provides a comparison between (i) a CNV detection method as disclosed herein ("current method", covering both aspects of sections A and B above) and (ii) a conventional CNV detection method ("previous method") in terms of CNV call accuracy (represented by the F measure) of the artificially contaminated sample and at three different total DNA input amounts (in nanograms-ng) subjected to array analysis and CNV detection. The proposed CNV detection method disclosed herein ("current method", covering both aspects of sections A and B above) provides improved accuracy for these artificial samples at all three input amount levels. It should be noted that samples of all contamination levels are included in the accuracy assessment here, but removing samples with the highest contamination level can lead to further improvements in accuracy levels.

[0068] Figure 6 Depicted are graphical representations of a comparison between (i) the CNV detection method as disclosed herein ("current method", covering both aspects of Sections A and B above) and (ii) a conventional CNV detection method ("previous method") in terms of CNV calling accuracy (represented by the F-measure) in detection of structural variant star alleles of the pharmacogenomics gene CYP2D6 across six sets, and demonstrating the improvement of such detection by the current method compared to the previous method.

[0069] therefore, Fig. 7A and Figure 7B Depicted are example processes according to aspects described herein. In one or more examples, these processes may be performed by a processor or processing circuit of one or more computer systems (such as those described herein). For example, implementing Fig. 7A and / or Figure 7B The codes or instructions of a process may be part of a module of a software / computer program.

[0070] Fig. 7A Depicts an example process for array-based targeted copy number variant detection. For example, this process can be used in the case of unknown / variable concentration samples. First refer to Fig. 7A , the process obtains (702) an intensity signal aggregate from an assay of a set of input samples comprising genetic material. In an example, the intensity signal aggregate is from a high throughput genotyping platform that genotypes the input samples using a microarray-based genotyping platform. The process continues with performing (704) a cross-sample calibration of the intensity signals of the intensity signal aggregate based on a reference sample.

[0071] Performing cross-sample calibration includes, for example, constructing a reference signal distribution based on the intensity signals of a reference sample, and also includes performing, for each input sample in the input sample set, (a) obtaining a corresponding intensity signal set in the intensity signal set corresponding to the input sample (wherein the intensity signal set corresponding to the input sample includes (i) a first subset C of intensity signals from the targeted genomic region of interest and (ii) a second subset B of intensity signals from genomic regions outside the targeted genomic region of interest), and (b) calibrating the intensity signals in C based on the reference signal distribution to generate a corresponding calibration intensity signal set corresponding to the input sample.

[0072] Calibration of the intensity signal in C corresponding to the set of intensity signals of the input sample may include constructing a mapping for the input sample based on the relationship between (i) the intensity signal in B and (ii) the reference signal distribution. As an example, establishing a mapping may include defining a mapping function M(x). As an example, M(x) may map the intensity signal x to: (i) for x present in B, M(x) = the matching intensity signal of the vector A of the reference signal intensity from the reference signal distribution, the reference signal intensity corresponding to the genomic region outside the region of interest of the targeted genome; (ii) for x not present in B but falling between multiple intensity signals in B, M(x) = linear interpolation of the M(x) mapping based on multiple intensity signals in B; and (iii) for x not present in B and not falling within the range of intensity signals in B, M(x) = extrapolation of the mapping based on the highest quantile and the lowest quantile in B.

[0073] In some examples, constructing a reference signal distribution calculates vector A as the cross-sample median of autosomal array probes outside of the targeted genomic region of interest.

[0074] In some examples, calibrating the intensity signals in C also includes mapping the intensity signals in C using a mapping function to produce a set of calibration intensity signals corresponding to the input sample.

[0075] Return to reference Fig. 7A , and based on the cross-sample calibration (704), the process continues: for each next input sample of the one or more input samples (706), and determining (708) at least one corresponding aggregated calibration signal from the targeted genomic region of interest from one or more corresponding calibration intensity signal sets corresponding to the one or more input samples. The determination can therefore produce an aggregated calibration signal set. The process then continues to detect (710) variants in the targeted genomic region of interest based on the aggregated calibration signal set. In the example, the variant is a copy number variant.

[0076] In some examples, obtaining the intensity signal aggregate includes correcting for contamination, such as described above and with reference to the following Figure 7B For example, in Fig. 7A In the context of a process of obtaining an intensity signal aggregate (702) may include, for a set of input samples, using a set of array hybridization control probes to identify probe hybridization bias by: (i) aggregating row-based normalized raw intensity values ​​from the control probes into an aggregate value c s , (ii) aggregating the row-based normalized intensity values ​​from assays targeting human genomic material into an aggregate value x s , and (iii) determine the contamination factor f s As x s and c s A function where f s 、x s and c s Determined from the input sample. In the example, the contamination factor f s The function is: f s =x s / c sThen, in embodiments, determining (708) at least one corresponding aggregation calibration signal for one or more input samples and based on one or more corresponding sets of calibration intensity signals corresponding to the one or more input samples (e.g., one aggregation calibration signal for each targeted region of the sample) includes, for at least one aggregation calibration signal: (i) determining a first aggregation signal from the sets of calibration intensity signals corresponding to the targeted regions of the input samples; and correcting the first aggregation signal using a contamination factor and generating a second aggregation signal, wherein the second aggregation signal is output as the aggregation calibration signal for the targeted regions of the input samples.

[0077] In an example, using the contamination coefficient and generating a second aggregated signal includes using a regression-based model to predict a contribution of contamination based on the contamination coefficient, determining a residual based on the first aggregated signal and the contribution of contamination predicted by the model, and determining a second aggregated signal based on the residual and a composite contamination coefficient from an input sample.

[0078] Thus, the input samples of the input sample set can contain at least one of (i) variable amounts or concentrations of DNA relative to each other, or (ii) different fractions of contaminant DNA relative to each other, and in addition, prior measurements of (i) DNA quantification of the input samples, (ii) normalization of the input samples, and (iii) the fraction or amount of DNA contaminants in the input samples are not known or required in order to perform the processes described herein and obtain accurate variant detection results.

[0079] Figure 7B An example process for signal correction based on contamination coefficients according to aspects described herein is depicted. Signal correction can be performed as described above, for example. Fig. 7A This may be performed as part of the described procedure or as a stand-alone calibration as an example. Figure 7B The process obtains (720) an aggregate of intensity signals from an assay of a set of input samples comprising genetic material. For example, the aggregate of intensity signals is from a high throughput genotyping platform that genotypes the input samples using a microarray-based genotyping platform. The process continues with (722) an array hybridization control probe set to identify probe hybridization bias by aggregating the row-based normalized raw intensity values ​​from the control probes into an aggregate value c s , the row-based normalized intensity values ​​from assays targeting human genomic material are aggregated into an aggregate value x s , and the pollution factor f s Determine x s and c s A function where f s 、x s and c sDetermined based on the input samples of the input sample set. In the example, the contamination factor f s The function is: f s =x s / c s .

[0080] Using the contamination coefficient, the process corrects (724) a first signal obtained based on the intensity signals of the set of intensity signals and generates a corrected signal. In an example, using the contamination coefficient and generating the corrected signal includes (i) using a regression-based model to predict the contribution of contamination based on the contamination coefficient, (ii) determining a residual based on the first signal and the contribution of contamination predicted by the model, and (iii) determining the corrected signal based on the residual and a composite contamination coefficient from the input sample. The first signal can be a first aggregated signal of the set of intensity signals from the set of intensity signals, wherein the first aggregated signal corresponds to a target region of an input sample of the set of input samples, and the corrected signal can be a corrected aggregated signal, for example, for the target region of the input sample.

[0081] The processes described herein may be performed by one or more computer systems, individually or collectively, such as one or more computer systems executing genome analysis software to perform various aspects described herein. Figure 8 Depicted are examples of computer systems and related devices that incorporate and / or use aspects described herein. Computer systems may also be referred to herein as data processing devices / systems, computing devices / systems / nodes, or simply computers. Computer systems may be based on one or more of a variety of system architectures and / or instruction set architectures, such as those provided by Intel Corporation (Santa Clara, California, USA) as examples. Figure 8A computer system 800 is shown communicating with an external device 812. The computer system 800 includes one or more processors 802, such as a central processing unit (CPU). The processor may include functional components used in instruction execution, such as obtaining program instructions from a location such as a cache or main memory, decoding program instructions and executing program instructions, accessing a memory for instruction execution, and writing the results of the executed instructions. The processor 802 may also include registers used by one or more functional components in the functional components. The computer system 800 also includes a memory 804, an input / output (I / O) device 808, and an I / O interface 810, which may be coupled to the processor 802 and to each other via one or more buses and / or other connections. The bus connection represents any one or more bus structures of several types of bus structures, including a memory bus or memory controller, a peripheral bus, an accelerated graphics port, and a processor or local bus using any bus architecture in a variety of bus architectures. As an example and not limitation, such architectures include an industry standard architecture (ISA), a microchannel architecture (MCA), an enhanced ISA (EISA), a video electronics standards association (VESA) local bus, and a peripheral component interconnect (PCI).

[0082] The memory 804 may be or include a main or system memory (e.g., random access memory) used in the execution of program instructions, a storage device such as a hard drive, flash media, or optical media as examples, and / or a cache memory as examples. The memory 804 may include, for example, a cache, such as a shared cache, which may be coupled to a local cache of the processor 802 (examples include an L1 cache, an L2 cache, etc.). Additionally, the memory 804 may be or include at least one computer program product having a set (e.g., at least one) of program modules, instructions, codes, etc., which are configured to perform the functions of the embodiments described herein when executed by one or more processors.

[0083] The memory 804 may store an operating system 805 and other computer programs 806, such as one or more computer programs / programs executed to perform the various aspects described herein. Specifically, the program / application may include computer-readable program instructions that may be configured to perform the functions of the embodiments of the various aspects described herein.

[0084] Examples of VO devices 808 include, but are not limited to, microphones, speakers, global positioning system (GPS) devices, cameras, lights, accelerometers, gyroscopes, magnetometers, sensor devices configured to sense light, proximity, heart rate, body and / or ambient temperature, blood pressure, and / or skin resistance, and activity monitors. The VO devices can be incorporated into the computer system as shown, although in some embodiments, the VO devices can be viewed as external devices (812) coupled to the computer system via one or more VO interfaces 810.

[0085] The computer system 800 may communicate with one or more external devices 812 via one or more VO interfaces 810. Exemplary external devices include keyboards, pointing devices, displays, and / or any other device that enables a user to interact with the computer system 800. Other exemplary external devices include any device that enables the computer system 800 to communicate with one or more other computing systems or peripherals, such as a printer. A network interface / adapter is an exemplary VO interface that enables the computer system 800 to communicate with one or more networks, such as a local area network (LAN), a general wide area network (WAN), and / or a public network (e.g., the Internet), thereby providing communication with other computing devices or systems, storage devices, etc. Ethernet-based interfaces (such as Wi-Fi) and The adapter is merely an example of a currently available type of network adapter used in computer systems (BLUETOOTH is a registered trademark of Bluetooth SIG, Inc., Kirkland, Washington, USA).

[0086] Communication between the VO interface 810 and the external device 812 may occur across a wired and / or wireless communication link 811, such as an Ethernet-based wired or wireless connection. Exemplary wireless connections include cellular, Wi-Fi, Proximity-based near-field or other types of wireless connections. More generally, communication link 811 may be any suitable wireless and / or wired communication link for transmitting data.

[0087] Specific external devices 812 may include one or more data storage devices that may store one or more programs, one or more computer-readable program instructions and / or data, etc. The computer system 800 may include and / or be coupled to and communicate with removable / non-removable, volatile / non-volatile computer system storage media (e.g., as an external device to the computer system). For example, the computer system may include and / or be coupled to a non-removable, non-volatile magnetic medium (commonly referred to as a "hard drive"), a disk drive for reading from and writing to a removable, non-volatile disk (e.g., a "floppy disk"), and / or an optical drive for reading from or writing to a removable, non-volatile optical disk such as a CD-ROM, DVD-ROM, or other optical media.

[0088] The computer system 800 can operate with numerous other general or special computing system environments or configurations. The computer system 800 can take any of a variety of forms, well-known examples of which include, but are not limited to, personal computer (PC) systems, server computer systems such as message servers, thin clients, fat clients, workstations, laptop computers, handheld devices, mobile devices / computers such as smart phones, tablet computers and wearable devices, multiprocessor systems, microprocessor-based systems, telephone devices, network devices (such as edge devices), virtualization devices, storage controllers, set-top boxes, programmable consumer electronics, network PCs, small computer systems, large computer systems, and distributed cloud computing environments including any of the above systems or devices, etc.

[0089] Aspects of the present invention may be systems, methods, and / or computer program products, any of which may be configured to perform or facilitate the aspects described herein.

[0090] In some embodiments, various aspects of the present invention can take the form of a computer program product, which can be embodied as a computer-readable medium. A computer-readable medium can be a tangible storage device / medium on which a computer-readable program code / instruction is stored. Exemplary computer-readable media include, but are not limited to, electronic, magnetic, optical or semiconductor storage devices or systems or any combination of the foregoing. Exemplary embodiments of computer-readable media include hard disk drives or other mass storage devices, electrical connections with wires, random access memories (RAMs), read-only memories (ROMs), erasable programmable read-only memories such as EPROMs or flash memories, optical fibers, portable computer disks / floppy disks such as compact disk read-only memories (CD-ROMs) or digital versatile disks (DVDs), optical storage devices, magnetic storage devices or any combination of the foregoing. Computer-readable media can be read by a processor, a processing unit, etc. to obtain data (e.g., instructions) from the medium for execution. In a specific example, a computer program product is or includes one or more computer-readable media, which include / store computer-readable program codes to provide and facilitate one or more aspects described herein.

[0091] As noted, the program instructions included or stored in / on a computer readable medium can be obtained and executed by any of the various suitable components such as a processor of a computer system, so that the computer system performs and operates in a particular manner. Such program instructions for performing operations to perform, realize or promote various aspects described herein can be written in any desired programming language, or compiled from the code written in any desired programming language. In some embodiments, such programming languages ​​include object-oriented and / or procedural programming languages ​​such as C, C++, C#, Java, etc.

[0092] Program code may include obtaining one or more program instructions for execution by one or more processors. Computer program instructions may be provided to one or more processors of, for example, one or more computer systems to produce a machine so that the program instructions perform, implement or facilitate various aspects of the present invention when executed by one or more processors, such as the actions or functions described in the flowcharts and / or block diagrams described herein. Therefore, in some embodiments, each frame or combination of frames of the flowchart illustrations and / or block diagrams depicted and described herein may be implemented by computer program instructions.

[0093] While various implementations are described above, these are merely examples.

[0094] The terms used herein are only for the purpose of describing specific embodiments and are not intended to be limiting. As used herein, unless the context clearly indicates otherwise, the singular forms "a", "an", and "the" are intended to also include plural forms. It should also be understood that when used in this specification, the terms "include" and / or "comprise" specify the presence of the features, integers, steps, operations, elements, and / or parts, but do not exclude the presence or addition of one or more other features, integers, steps, operations, elements, parts, and / or groups thereof.

[0095] The corresponding structures, materials, actions, and equivalents of all means or steps plus functional elements in the following claims, if any, are intended to include any structure, material, or action for performing the function in combination with other claimed elements specifically claimed. The description of one or more embodiments has been presented for the purposes of illustration and description, but is not intended to be exhaustive or limited to the disclosed forms. Many modifications and variations will be apparent to those of ordinary skill in the art. The embodiments are selected and described in order to best explain the various aspects and practical applications, and to enable others of ordinary skill in the art to understand the various embodiments with various modifications suitable for the specific purposes contemplated.

Claims

1. A computer-implemented method comprising: obtaining an aggregate of intensity signals from measurements of a set of input samples comprising genetic material; Performing a cross-sample calibration on the intensity signal of the intensity signal aggregate based on one or more reference samples, wherein performing the cross-sample calibration comprises: constructing a reference signal distribution based on the intensity signals of the one or more reference samples; and For one or more input samples of the input sample set: Obtaining a corresponding intensity signal set corresponding to the input sample in the intensity signal set, the intensity signal set corresponding to the input sample comprising (i) a first subset C of intensity signals from one or more targeted genomic regions of interest and (ii) a second subset B of intensity signals from at least one genomic region other than the one or more targeted genomic regions of interest; and calibrating the intensity signals in C based on the reference signal distribution to generate a corresponding set of calibration intensity signals corresponding to the input sample; determining, for the one or more input samples and based on one or more corresponding sets of calibration intensity signals corresponding to the one or more input samples, at least one corresponding aggregated calibration signal from the one or more targeted genomic regions of interest, wherein the determining produces an aggregated calibration signal set; and One or more variants in the one or more targeted genomic regions of interest are detected based on the aggregate calibration signal collection.

2. A method according to claim 1, wherein the calibration of the intensity signal in C of the intensity signal set corresponding to the input sample includes establishing a mapping for the input sample based on (i) the relationship between the intensity signal in B and (ii) the reference signal distribution.

3. The method according to claim 2, wherein said establishing said mapping comprises defining a mapping function M(x) such that M(x) maps the intensity signal x to: For x present in B, M(x) = the matching intensity signal of the vector A of reference signal intensities from the reference signal distribution, the reference signal intensities corresponding to at least one genomic region outside of the one or more targeted genomic regions of interest; For x that does not exist in B but falls between multiple intensity signals in B, M(x)=linear interpolation of the M(x) mapping based on the multiple intensity signals in B; and For x that is not present in B and does not fall within the intensity signal range in B, M(x) = extrapolation of the mapping based on the highest and lowest quantiles in B.

4. The method of claim 3, wherein said constructing said reference signal distribution calculates said vector A as a cross-sample median of autosomal array probes outside of said one or more targeted genomic regions of interest.

5. The method according to claim 3 or 4, wherein the calibrating the intensity signals in C also includes using the mapping function to map the intensity signals in C to generate the calibration intensity signal set corresponding to the input sample.

6. The method of claim 1, 2, 3, 4 or 5, wherein obtaining the intensity signal aggregate comprises, for the input sample set, identifying probe hybridization bias using an array hybridization control probe set by aggregating row-based normalized raw intensity values ​​from the control probes into an aggregate value c s , the row-based normalized intensity values ​​from assays targeting human genomic material are aggregated into an aggregate value x s , and the pollution factor f s Determine x s and c s function, where f s 、x s and c s Determined based on input sample.

7. The method according to claim 6, wherein the contamination factor f s The function is: f s =x s / c s .

8. The method of claim 6 or 7, wherein for the one or more input samples and based on one or more corresponding sets of calibration intensity signals corresponding to the one or more input samples, determining at least one corresponding aggregate calibration signal comprises for the at least one aggregate calibration signal: determining a first aggregation signal from a set of calibrated intensity signals corresponding to a targeted region of the input sample; and The contamination coefficient is used to correct the first aggregation signal and generate a second aggregation signal, wherein the second aggregation signal is output as the aggregation calibration signal for the targeted region of the input sample.

9. A method according to claim 8, wherein using the contamination coefficient and generating the second aggregate signal includes (i) using a regression-based model to predict the contribution of contamination based on the contamination coefficient, (ii) determining a residual based on the first aggregate signal and the contribution of contamination predicted by the model, and (iii) determining the second aggregate signal based on the residual and a composite contamination coefficient from the input sample.

10. The method of claim 1, 2, 3, 4, 5, 6, 7, 8 or 9, wherein the one or more variants are one or more copy number variants.

11. The method of claim 1, 2, 3, 4, 5, 6, 7, 8, 9 or 10, wherein (i) deoxyribonucleic acid (DNA) quantification of the input sample, (ii) normalization of the input sample, and (iii) prior measurement of the fraction or amount of DNA contaminants in the input sample are not known or required in performing the method.

12. A method according to claims 1, 2, 3, 4, 5, 6, 7, 8, 9, 10 or 11, wherein the input samples of the input sample set contain at least one of (i) variable amounts or concentrations of deoxyribonucleic acid (DNA) relative to each other or (ii) different fractions of contaminant DNA relative to each other.

13. The method of claim 1, 2, 3, 4, 5, 6, 7, 8, 9, 10, 11 or 12, wherein the intensity signal aggregate is from a high throughput genotyping platform that genotypes the input sample using a microarray-based genotyping platform.

14. A computer system, comprising: Memory; as well as a processor in communication with the memory, wherein the computer system is configured to perform a method comprising the steps of: obtaining an aggregate of intensity signals from measurements of a set of input samples comprising genetic material; Performing a cross-sample calibration on the intensity signal of the intensity signal aggregate based on one or more reference samples, wherein performing the cross-sample calibration comprises: constructing a reference signal distribution based on the intensity signals of the one or more reference samples; and For one or more input samples of the input sample set: Obtaining a corresponding intensity signal set corresponding to the input sample in the intensity signal set, the intensity signal set corresponding to the input sample comprising (i) a first subset C of intensity signals from one or more targeted genomic regions of interest and (ii) a second subset B of intensity signals from at least one genomic region other than the one or more targeted genomic regions of interest; and calibrating the intensity signals in C based on the reference signal distribution to generate a corresponding set of calibration intensity signals corresponding to the input sample; determining, for the one or more input samples and based on one or more corresponding sets of calibration intensity signals corresponding to the one or more input samples, at least one corresponding aggregated calibration signal from the one or more targeted genomic regions of interest, wherein the determining produces an aggregated calibration signal set; and One or more variants in the one or more targeted genomic regions of interest are detected based on the aggregate calibration signal collection.

15. A computer system according to claim 14, wherein the calibration of the intensity signal in C of the intensity signal set corresponding to the input sample includes establishing a mapping for the input sample based on the relationship between (i) the intensity signal in B and (ii) the reference signal distribution.

16. The computer system of claim 15, wherein said establishing said mapping comprises defining a mapping function M(x) such that M(x) maps the intensity signal x to: For x present in B, M(x) = the matching intensity signal of the vector A of reference signal intensities from the reference signal distribution, the reference signal intensities corresponding to at least one genomic region outside of the one or more targeted genomic regions of interest; For x that does not exist in B but falls between multiple intensity signals in B, M(x)=linear interpolation of the M(x) mapping based on the multiple intensity signals in B; and For x that is not present in B and does not fall within the intensity signal range in B, M(x) = extrapolation of the mapping based on the highest and lowest quantiles in B.

17. The computer system of claim 16, wherein said constructing said reference signal distribution calculates said vector A as a cross-sample median of autosomal array probes outside of said one or more targeted genomic regions of interest.

18. The computer system of claim 16 or 17, wherein the calibrating the intensity signals in C further comprises mapping the intensity signals in C using the mapping function to generate the set of calibration intensity signals corresponding to the input samples.

19. The computer system of claim 14, 15, 16, 17 or 18, wherein said obtaining said intensity signal aggregate comprises, for said input sample set, using an array hybridization control probe set to identify probe hybridization bias by: aggregating row-based normalized raw intensity values ​​from said control probes into an aggregate value c s , the row-based normalized intensity values ​​from assays targeting human genomic material are aggregated into an aggregate value x s , and the pollution factor f s Determine x s and c s function, where f s 、x s and c s Determined based on input sample.

20. The computer system of claim 19, wherein the contamination factor f s The function is: f s =x s / c s 。 21. The computer system of claim 19 or 20, wherein for the one or more input samples and based on the one or more sets of corresponding calibration intensity signals corresponding to the one or more input samples, determining the at least one corresponding aggregated calibration signal comprises for an aggregated calibration signal in the at least one aggregated calibration signal: determining a first aggregation signal from a set of calibrated intensity signals corresponding to a targeted region of the input sample; and The contamination coefficient is used to correct the first aggregation signal and generate a second aggregation signal, wherein the second aggregation signal is output as the aggregation calibration signal for the targeted region of the input sample.

22. A computer system according to claim 21, wherein the use of the contamination coefficient and generating the second aggregate signal includes (i) using a regression-based model to predict the contribution of contamination based on the contamination coefficient, (ii) determining a residual based on the first aggregate signal and the contribution of contamination predicted by the model, and (iii) determining the second aggregate signal based on the residual and a composite contamination coefficient from the input sample.

23. The computer system of claim 14, 15, 16, 17, 18, 19, 20, 21 or 22, wherein the one or more variants are one or more copy number variants.

24. A computer system according to claims 14, 15, 16, 17, 18, 19, 20, 21, 22 or 23, wherein (i) deoxyribonucleic acid (DNA) quantification of the input sample, (ii) normalization of the input sample and (iii) previous measurement of the fraction or amount of DNA contaminants in the input sample are not known or required in performing the method.

25. A computer system according to claims 14, 15, 16, 17, 18, 19, 20, 21, 22, 23 or 24, wherein the input samples of the input sample set contain at least one of (i) variable amounts or concentrations of deoxyribonucleic acid (DNA) relative to each other or (ii) different fractions of contaminant DNA relative to each other.

26. The computer system of claim 14, 15, 16, 17, 18, 19, 20, 21, 22, 23, 24 or 25, wherein the collection of intensity signals is from a high throughput genotyping platform that genotypes the input sample using a microarray-based genotyping platform.

27. A computer program product, the computer program product comprising: A computer-readable storage medium capable of being read by a processing circuit and storing instructions, the instructions being executed by the processing circuit for performing a method, the method comprising: obtaining an aggregate of intensity signals from measurements of a set of input samples comprising genetic material; Performing a cross-sample calibration on the intensity signal of the intensity signal aggregate based on one or more reference samples, wherein performing the cross-sample calibration comprises: constructing a reference signal distribution based on the intensity signals of the one or more reference samples; and For one or more input samples of the input sample set: Obtaining a corresponding intensity signal set corresponding to the input sample in the intensity signal set, the intensity signal set corresponding to the input sample comprising (i) a first subset C of intensity signals from one or more targeted genomic regions of interest and (ii) a second subset B of intensity signals from at least one genomic region other than the one or more targeted genomic regions of interest; and calibrating the intensity signals in C based on the reference signal distribution to generate a corresponding set of calibration intensity signals corresponding to the input sample; determining, for the one or more input samples and based on one or more corresponding sets of calibration intensity signals corresponding to the one or more input samples, at least one corresponding aggregated calibration signal from the one or more targeted genomic regions of interest, wherein the determining produces an aggregated calibration signal set; and One or more variants in the one or more targeted genomic regions of interest are detected based on the aggregate calibration signal collection.

28. A computer program product according to claim 27, wherein the calibration of the intensity signal in C of the set of intensity signals corresponding to the input sample includes establishing a mapping for the input sample based on the relationship between (i) the intensity signal in B and (ii) the reference signal distribution.

29. The computer program product of claim 28, wherein said establishing said mapping comprises defining a mapping function M(x) such that M(x) maps an intensity signal x to: For x present in B, M(x) = the matching intensity signal of the vector A of reference signal intensities from the reference signal distribution, the reference signal intensities corresponding to at least one genomic region outside of the one or more targeted genomic regions of interest; For x that does not exist in B but falls between multiple intensity signals in B, M(x)=linear interpolation of the M(x) mapping based on the multiple intensity signals in B; and For x that is not present in B and does not fall within the intensity signal range in B, M(x) = extrapolation of the mapping based on the highest and lowest quantiles in B.

30. The computer program product of claim 29, wherein said constructing said reference signal distribution calculates said vector A as a cross-sample median of autosomal array probes outside of said one or more targeted genomic regions of interest.

31. A computer program product according to claim 29 or 30, wherein the calibrating the intensity signals in C further comprises mapping the intensity signals in C using the mapping function to produce the set of calibration intensity signals corresponding to the input samples.

32. The computer program product of claim 27, 28, 29, 30, or 31, wherein said obtaining said intensity signal aggregate comprises, for said input sample set, identifying probe hybridization bias using a set of array hybridization control probes by aggregating row-based normalized raw intensity values ​​from said control probes into an aggregate value c s , the row-based normalized intensity values ​​from assays targeting human genomic material are aggregated into an aggregate value x s , and the pollution factor f s Determine x s and c s function, where f s 、x s and c s Determined based on input sample.

33. The computer program product of claim 32, wherein the contamination factor f s The function is: f s =x s / c s .

34. The computer program product of claim 32 or 33, wherein for the one or more input samples and based on the one or more sets of corresponding calibration intensity signals corresponding to the one or more input samples, determining the at least one corresponding aggregated calibration signal comprises, for an aggregated calibration signal in the at least one aggregated calibration signal: determining a first aggregation signal from a set of calibrated intensity signals corresponding to a targeted region of the input sample; and The contamination coefficient is used to correct the first aggregation signal and generate a second aggregation signal, wherein the second aggregation signal is output as the aggregation calibration signal for the targeted region of the input sample.

35. A computer program product according to claim 34, wherein the use of the contamination coefficient and generating the second aggregate signal includes (i) using a regression-based model to predict the contribution of contamination based on the contamination coefficient, (ii) determining a residual based on the first aggregate signal and the contribution of contamination predicted by the model, and (iii) determining the second aggregate signal based on the residual and a composite contamination coefficient from the input sample.

36. The computer program product of claim 27, 28, 29, 30, 31, 32, 33, 34 or 35, wherein the one or more variants are one or more copy number variants.

37. A computer program product according to claims 27, 28, 29, 30, 31, 32, 33, 34, 35 or 36, wherein (i) deoxyribonucleic acid (DNA) quantification of the input sample, (ii) normalization of the input sample and (iii) previous measurement of the fraction or amount of DNA contaminants in the input sample are not known or required in performing the method.

38. A computer program product according to claims 27, 28, 29, 30, 31, 32, 33, 34, 35, 36 or 37, wherein the input samples of the input sample set contain at least one of (i) variable amounts or concentrations of deoxyribonucleic acid (DNA) relative to each other or (ii) different fractions of contaminant DNA relative to each other.

39. The computer program product of claim 27, 28, 29, 30, 31, 32, 33, 34, 35, 36, 37 or 38, wherein the aggregate of intensity signals is from a high throughput genotyping platform that genotypes the input sample using a microarray-based genotyping platform.

40. A computer-implemented method comprising: obtaining an aggregate of intensity signals from measurements of a set of input samples comprising genetic material; Probe hybridization bias is identified using an array hybridization control probe set by aggregating the row-based normalized raw intensity values ​​from the control probes into an aggregate value c s , the row-based normalized intensity values ​​from assays targeting human genomic material are aggregated into an aggregate value x s , and the pollution factor is determined as x s and c s function, where f s 、x s and c s determining according to an input sample of the input sample set; as well as The contamination coefficient is used to correct a first signal obtained based on the intensity signal of the intensity signal aggregate, and a correction signal is generated.

41. The method of claim 40, wherein using the contamination coefficient and generating the correction signal comprises (i) using a regression-based model to predict the contribution of contamination based on the contamination coefficient, (ii) determining a residual based on the first signal and the contribution of contamination predicted by the model, and (iii) determining the correction signal based on the residual and a composite contamination coefficient from the input sample.

42. The method according to claim 40 or 41, wherein the contamination factor f s The function is: f s =x s / c s 。 43. A method according to claim 40, 41 or 42, wherein the first signal is a first aggregate signal from a set of intensity signals of the intensity signal set, the first aggregate signal corresponds to a target area of ​​an input sample of the input sample set, and wherein the correction signal is a corrected aggregate signal.

44. The method of claim 40, 41, 42 or 43, wherein the aggregate of intensity signals is from a high throughput genotyping platform that genotypes the input sample using a microarray-based genotyping platform.

45. A computer system, the computer system comprising: Memory; as well as a processor in communication with the memory, wherein the computer system is configured to perform a method comprising the steps of: obtaining an aggregate of intensity signals from measurements of a set of input samples comprising genetic material; Probe hybridization bias is identified using an array hybridization control probe set by aggregating the row-based normalized raw intensity values ​​from the control probes into an aggregate value c s , the row-based normalized intensity values ​​from assays targeting human genomic material are aggregated into an aggregate value x s , and the pollution factor is determined as x s and c s function, where f s 、x s and c s determining according to an input sample of the input sample set; as well as The contamination coefficient is used to correct a first signal obtained based on the intensity signal of the intensity signal aggregate, and a correction signal is generated.

46. ​​A computer system according to claim 45, wherein using the contamination coefficient and generating the correction signal includes (i) using a regression-based model to predict the contribution of contamination based on the contamination coefficient, (ii) determining a residual based on the first signal and the contribution of contamination predicted by the model, and (iii) determining the correction signal based on the residual and a composite contamination coefficient from the input sample.

47. A computer system according to claim 45 or 46, wherein the contamination factor f s The function is: f s =x s / c s .

48. A computer system according to claim 45, 46 or 47, wherein the first signal is a first aggregate signal from a set of intensity signals of the intensity signal set, the first aggregate signal corresponds to a target area of ​​an input sample of the input sample set, and wherein the correction signal is a corrected aggregate signal.

49. The computer system of claim 45, 46, 47 or 48, wherein the aggregate of intensity signals is from a high throughput genotyping platform that genotypes the input sample using a microarray-based genotyping platform.

50. A computer program product, the computer program product comprising: A computer-readable storage medium capable of being read by a processing circuit and storing instructions, the instructions being executed by the processing circuit for performing a method, the method comprising: obtaining an aggregate of intensity signals from measurements of a set of input samples comprising genetic material; Probe hybridization bias is identified using an array hybridization control probe set by aggregating the row-based normalized raw intensity values ​​from the control probes into an aggregate value c s , the row-based normalized intensity values ​​from assays targeting human genomic material are aggregated into an aggregate value x s , and the pollution factor is determined as x s and c s function, where f s 、x s and c s Determined based on input samples of the input sample set; and The contamination coefficient is used to correct a first signal obtained based on the intensity signal of the intensity signal aggregate, and a correction signal is generated.

51. A computer program product according to claim 50, wherein using the contamination coefficient and generating the correction signal includes (i) using a regression-based model to predict the contribution of contamination based on the contamination coefficient, (ii) determining a residual based on the first signal and the contribution of contamination predicted by the model, and (iii) determining the correction signal based on the residual and a composite contamination coefficient from the input sample.

52. A computer program product according to claim 50 or 51, wherein the contamination factor f s The function is: f s =x s / c s .

53. A computer program product according to claim 50, 51 or 52, wherein the first signal is a first aggregate signal from the set of intensity signals of the set of intensity signals, the first aggregate signal corresponds to a target area of ​​an input sample of the input sample set, and wherein the correction signal is a corrected aggregate signal.

54. The computer program product of claim 50, 51, 52 or 53, wherein the aggregate of intensity signals is from a high throughput genotyping platform that genotypes the input sample using a microarray-based genotyping platform.

Images