Methods and systems for refining copy number variation in a liquid biopsy assay

Abstract

Methods, systems, and software are provided for validating a copy number variation in a test subject. A first dataset is obtained comprising bin-level sequence ratios, segment-level sequence ratios and segment-level measures of dispersion. Bins representing regions of a human reference genome are determined from sequencing cell-free nucleic acids in a liquid biopsy sample and reference samples. Segments encompass subsets of adjacent bins, where segment-level sequence ratios and measures of dispersion are determined using bin-level sequence ratios. A copy number status annotation for a segment is validated by applying the first dataset to a plurality of filters comprising a measure of central tendency bin-level sequence ratio filter, a confidence filter, and a measure of central tendency-plus-deviation bin-level sequence ratio filter. When a filter is fired, the copy number status annotation of the segment is rejected; and when no filter is fired, the copy number status annotation of the segment is validated.

1 - 109 . (canceled) 
     
     
         110 . A method of validating a copy number variation status for a test subject, the method comprising:
 at a computer system having one or more processors, and memory storing one or more programs for execution by the one or more processors:   (A) aligning each respective nucleic acid sequence in a first plurality of at least 10,000 nucleic acid sequences obtained by sequencing cell-free DNA fragments in a liquid biopsy sample of the test subject to a reference construct, wherein the reference construct represents at least 25 Mb of the genome for the species of the subject; and   (B) determining:
 (1) a plurality of at least 500 bin-level values using the plurality of at least 10,000 nucleic acid sequences, each respective bin-level value in the plurality of at least 500 bin-level values corresponding to a respective bin in a plurality of at least 500 bins, wherein:
 each respective bin in the plurality of at least 500 bins represents a corresponding region of a reference genome for the species of the subject, 
 the plurality of at least 500 bins collectively covers at least 25 Mb of the genome for the species of the test subject, and 
 each respective bin-level value in the plurality of at least 500 bin-level values is determined from a comparison between (i) a proportion of nucleic acid sequences in the first plurality of at least 10,000 nucleic acid sequences that map to the corresponding bin in the plurality of at least 500 bins, and (ii) a proportion of nucleic acid sequences for one or more reference samples that map to the corresponding bin in the plurality of at least 500 bins; 
 
 (2) a plurality of segment-level values using the plurality of at least 500 bin-level values, each respective segment-level value in the plurality of segment-level values corresponding to a segment in a plurality of segments, wherein:
 each respective segment in the plurality of segments represents a corresponding region of the reference genome for the species of the subject encompassing a corresponding subset of adjacent bins in the plurality of bins, and 
 each respective segment-level value in the plurality of segment-level values is determined from a measure of central tendency of the plurality of bin-level values corresponding to the subset of adjacent bins encompassed by the respective segment; and 
 
 (3) a plurality of segment-level measures of dispersion using the plurality of at least 500 bin-level values, each respective segment-level measure of dispersion in the plurality of segment-level measures of dispersion corresponding to a respective segment in the plurality of segments, wherein:
 each respective segment-level measure of dispersion in the plurality of segment-level measures of dispersion is determined from the plurality of bin-level values corresponding to the subset of adjacent bins encompassed by the respective segment, 
 
   wherein the copy number variation status of a respective segment in the plurality of segments is validated when one or more conditions are satisfied, the one or more conditions comprising that the segment-level measure of dispersion corresponding to the respective segment satisfies one or more confidence thresholds.   
     
     
         111 . The method of  claim 110 , wherein each respective segment-level measure of dispersion in the plurality of segment-level measures of dispersion is a confidence interval, a standard deviation, a standard error, a variance, or a range. 
     
     
         112 . The method of  claim 110 , wherein
 each respective segment-level measure of dispersion in the plurality of segment-level measures of dispersion is a confidence interval, and   determining each respective segment-level measure of dispersion in the plurality of segment-level measures of dispersion comprises bootstrapping the plurality of bin-level values corresponding to the subset of adjacent bins encompassed by the respective segment.   
     
     
         113 . The method of  claim 110 , wherein the respective segment-level measure of dispersion corresponding to the respective segment fails to satisfy a respective confidence threshold in the one or more confidence thresholds when a lower bound of the respective measure of dispersion is lower than the respective confidence threshold. 
     
     
         114 . The method of  claim 110 , wherein the respective segment-level measure of dispersion corresponding to the respective segment fails to satisfy a respective confidence threshold in the one or more confidence thresholds when an upper bound of the respective measure of dispersion is higher than the respective confidence threshold. 
     
     
         115 . The method of  claim 110 , wherein a respective confidence threshold in the one or more confidence thresholds is a measure of central tendency of the segment-level values corresponding to all other segments that map to the same chromosome of the reference genome for the species of the subject as the respective segment. 
     
     
         116 . The method of  claim 110 , wherein each respective bin-level value in the plurality of bin-level values is derived from a comparison of (a) a test read depth of nucleic acid sequences, in the plurality of at least 10,000 nucleic acid sequences, that map to the corresponding bin in the plurality of bins, to (b) a measure of central tendency of reference nucleic acid sequence depths of nucleic acid sequences, for each reference sample of a plurality of reference samples, that map to the corresponding bin. 
     
     
         117 . The method of  claim 110 , wherein the one or more conditions further comprise a requirement that a measure of central tendency of the plurality of bin-level values corresponding to the subset of bins encompassed by the respective segment satisfies one or more bin-level thresholds. 
     
     
         118 . The method of  claim 117 , wherein the measure of central tendency of the plurality of bin-level values corresponding to the subset of bins encompassed by the respective segment satisfies the one or more bin-level thresholds when the measure of central tendency is greater than a bin-level amplification threshold. 
     
     
         119 . The method of  claim 117 , wherein the measure of central tendency of the plurality of bin-level values corresponding to the subset of bins encompassed by the respective segment satisfies the one or more bin-level thresholds when the measure of central tendency is less than a bin-level deletion threshold. 
     
     
         120 . The method of  claim 110 , the method further comprising:
 using circular binary segmentation (CBS) to identify the subset of adjacent bins encompassed by the respective segment are through their similarity in respective bin-level values.   
     
     
         121 . The method of  claim 110 , wherein the measure of central tendency of the plurality of bin-level values corresponding to the subset of adjacent bins encompassed by the respective segment in the (A) obtaining is a weighted mean. 
     
     
         122 . The method of  claim 110 , wherein the one or more conditions further comprise a requirement that a measure of central tendency of the plurality of bin-level values corresponding to the subset of bins encompassed by the respective segment satisfies one or more measure of central tendency-plus-deviation bin-level thresholds, wherein the one or more measure of central tendency-plus-deviation bin-level thresholds are derived from (i) a measure of central tendency of the bin-level values corresponding to the plurality of bins that map to the same chromosome of the reference genome for the species as the respective segment, and (ii) a measure of dispersion across the bin-level values corresponding to the plurality of bins that map to the same chromosome of the reference genome for the species of the subject as the respective segment. 
     
     
         123 . The method of  claim 122 , wherein:
 a respective measure of central tendency-plus-deviation bin-level threshold in the one or more measure of central tendency-plus-deviation bin-level thresholds is a sum of:
 (i) a measure of central tendency of the bin-level values corresponding to the plurality of bins that map to the same chromosome, and 
 (ii) a measure of central tendency of a plurality of absolute dispersions, wherein:
 each absolute dispersion is determined using a comparison between each respective bin-level value corresponding to each respective bin in the plurality of bins that map to the same chromosome as the respective segment, and the measure of central tendency of the bin-level values measured in (i); and 
 
   the measure of central tendency of the plurality of bin-level values corresponding to the subset of bins encompassed by the respective segment fails to satisfy the one or more measure of central tendency-plus-deviation bin-level thresholds when the measure of central tendency of the plurality of bin-level values corresponding to the subset of bins encompassed by the respective segment is less than the respective measure of central tendency-plus-deviation bin-level threshold.   
     
     
         124 . The method of  claim 122 , wherein:
 a respective measure of central tendency-plus-deviation bin-level threshold in the one or more measure of central tendency-plus-deviation bin-level thresholds comprises:
 (i) a measure of central tendency of the bin-level values corresponding to the plurality of bins that map to the same chromosome, minus 
 (ii) a measure of central tendency of a plurality of absolute dispersions, wherein:
 each absolute dispersion is determined using a comparison between each respective bin-level value corresponding to each respective bin in the plurality of bins that map to the same chromosome as the respective segment, and the measure of central tendency of the bin-level values measured in (i); and 
 
   the measure of central tendency of the plurality of bin-level values corresponding to the subset of bins encompassed by the respective segment fails to satisfy the one or more measure of central tendency-plus-deviation bin-level thresholds when the measure of central tendency of the plurality of bin-level values corresponding to the subset of bins encompassed by the respective segment is greater than the respective measure of central tendency-plus-deviation bin-level threshold.   
     
     
         125 . The method of  claim 124 , wherein the respective measure of central tendency-plus-deviation bin-level threshold in the one or more measure of central tendency-plus-deviation bin-level sequence ratio thresholds is:
 (i) the measure of central tendency of the bin-level values corresponding to the plurality of bins that map to the same chromosome as the respective segment, minus   (ii) the measure of central tendency of the plurality of absolute dispersions multiplied by a factor k.   
     
     
         126 . The method of  claim 110 , wherein the one or more conditions further comprise a requirement that the segment-level value corresponding to the respective segment satisfies one or more segment-level thresholds. 
     
     
         127 . The method of  claim 126 , wherein the segment-level value corresponding to the respective segment satisfies the one or more segment-level thresholds when the segment-level value is greater than a segment-level amplification threshold. 
     
     
         128 . The method of  claim 126 , wherein the segment-level value corresponding to the respective segment satisfies the one or more segment-level thresholds when the segment-level value is less than a segment-level deletion threshold. 
     
     
         129 . A computer system comprising:
 one or more processors; and   a non-transitory computer-readable medium including computer-executable instructions that, when executed by the one or more processors, cause the processors to perform a method comprising:   (A) aligning each respective nucleic acid sequence in a first plurality of at least 10,000 nucleic acid sequences obtained by sequencing cell-free DNA fragments in a liquid biopsy sample of the test subject to a reference construct, wherein the reference construct represents at least 25 Mb of the genome for the species of the subject; and   (B) determining:
 (1) a plurality of at least 500 bin-level values using the plurality of at least 10,000 nucleic acid sequences, each respective bin-level value in the plurality of at least 500 bin-level values corresponding to a respective bin in a plurality of at least 500 bins, wherein:
 each respective bin in the plurality of at least 500 bins represents a corresponding region of a reference genome for the species of the subject, 
 the plurality of at least 500 bins collectively covers at least 25 Mb of the genome for the species of the test subject, and 
 each respective bin-level value in the plurality of at least 500 bin-level values is determined from a comparison between (i) a proportion of nucleic acid sequences in the first plurality of at least 10,000 nucleic acid sequences that map to the corresponding bin in the plurality of at least 500 bins, and (ii) a proportion of nucleic acid sequences for one or more reference samples that map to the corresponding bin in the plurality of at least 500 bins; 
 
 (2) a plurality of segment-level values using the plurality of at least 500 bin-level values, each respective segment-level value in the plurality of segment-level values corresponding to a segment in a plurality of segments, wherein:
 each respective segment in the plurality of segments represents a corresponding region of the reference genome for the species of the subject encompassing a corresponding subset of adjacent bins in the plurality of bins, and 
 each respective segment-level value in the plurality of segment-level values is determined from a measure of central tendency of the plurality of bin-level values corresponding to the subset of adjacent bins encompassed by the respective segment; and 
 
 (3) a plurality of segment-level measures of dispersion using the plurality of at least 500 bin-level values, each respective segment-level measure of dispersion in the plurality of segment-level measures of dispersion corresponding to a respective segment in the plurality of segments, wherein:
 each respective segment-level measure of dispersion in the plurality of segment-level measures of dispersion is determined from the plurality of bin-level values corresponding to the subset of adjacent bins encompassed by the respective segment, 
 
   
       wherein the copy number variation status of a respective segment in the plurality of segments is validated when one or more conditions are satisfied, the one or more conditions comprising that the segment-level measure of dispersion corresponding to the respective segment satisfies one or more confidence thresholds. 
     
     
         130 . A non-transitory computer-readable storage medium having stored thereon program code instructions that, when executed by a processor, cause the processor to perform a method comprising:
 (A) aligning each respective nucleic acid sequence in a first plurality of at least 10,000 nucleic acid sequences obtained by sequencing cell-free DNA fragments in a liquid biopsy sample of the test subject to a reference construct, wherein the reference construct represents at least 25 Mb of the genome for the species of the subject; and   (B) determining:
 (1) a plurality of at least 500 bin-level values using the plurality of at least 10,000 nucleic acid sequences, each respective bin-level value in the plurality of at least 500 bin-level values corresponding to a respective bin in a plurality of at least 500 bins, wherein:
 each respective bin in the plurality of at least 500 bins represents a corresponding region of a reference genome for the species of the subject, 
 the plurality of at least 500 bins collectively covers at least 25 Mb of the genome for the species of the test subject, and 
 each respective bin-level value in the plurality of at least 500 bin-level values is determined from a comparison between (i) a proportion of nucleic acid sequences in the first plurality of at least 10,000 nucleic acid sequences that map to the corresponding bin in the plurality of at least 500 bins, and (ii) a proportion of nucleic acid sequences for one or more reference samples that map to the corresponding bin in the plurality of at least 500 bins; 
 
 (2) a plurality of segment-level values using the plurality of at least 500 bin-level values, each respective segment-level value in the plurality of segment-level values corresponding to a segment in a plurality of segments, wherein:
 each respective segment in the plurality of segments represents a corresponding region of the reference genome for the species of the subject encompassing a corresponding subset of adjacent bins in the plurality of bins, and 
 each respective segment-level value in the plurality of segment-level values is determined from a measure of central tendency of the plurality of bin-level values corresponding to the subset of adjacent bins encompassed by the respective segment; and 
 
 (3) a plurality of segment-level measures of dispersion using the plurality of at least 500 bin-level values, each respective segment-level measure of dispersion in the plurality of segment-level measures of dispersion corresponding to a respective segment in the plurality of segments, wherein:
 each respective segment-level measure of dispersion in the plurality of segment-level measures of dispersion is determined from the plurality of bin-level values corresponding to the subset of adjacent bins encompassed by the respective segment, 
 
   
       wherein the copy number variation status of a respective segment in the plurality of segments is validated when one or more conditions are satisfied, the one or more conditions comprising that the segment-level measure of dispersion corresponding to the respective segment satisfies one or more confidence thresholds.