Size-based analysis of cell-free tumor dna for classifying level of cancer

Abstract

A fractional concentration of tumor DNA in a plasma sample is estimated by analyzing a biological sample of an organism. One or more sequence reads obtained from a sequencing of the DNA fragment are received. The one or more sequence reads are aligned to a reference genome to obtain aligned locations for both ends of the DNA fragment. Using the aligned locations, a size of the DNA fragment is determined. For each size of a plurality of sizes, an amount of a set of the plurality of DNA fragments from the plasma sample corresponding to the size is determined. A first value of a first parameter is calculated based on the amounts of DNA fragments at multiple sizes. The first value is compared to a calibration value. A fractional concentration of tumor DNA in the plasma sample is estimated based on the comparison.

What is claimed is: 
     
         1 . A method of analyzing a plasma sample of an organism, the plasma sample including cell-free DNA fragments originating from normal cells and potentially from cells associated with cancer, the method comprising:
 for each of a plurality of DNA fragments from the plasma sample:
 receiving one or more sequence reads obtained from a sequencing of the DNA fragment, the one or more sequence reads including both ends of the DNA fragment; 
 aligning the one or more sequence reads to a reference genome to obtain aligned locations for both ends of the DNA fragment; and 
 using the aligned locations to determine a size of the DNA fragment; 
   for each size of a plurality of sizes:
 determining an amount of a set of the plurality of DNA fragments from the plasma sample corresponding to the size, using the sizes determined from the aligned locations for the set of the plurality of DNA fragments; 
   calculating a first value of a first parameter based on the amounts of DNA fragments at multiple sizes, the first parameter providing a statistical measure of a size profile of DNA fragments in the plasma sample;   comparing the first value to a calibration value; and   estimating a fractional concentration of tumor DNA in the plasma sample based on the comparison.   
     
     
         2 . The method of  claim 1 , wherein the DNA fragments are chosen at random. 
     
     
         3 . The method of  claim 1 , wherein the DNA fragments are chosen from more than one region of a genome. 
     
     
         4 . The method of  claim 3 , wherein the DNA fragments are chosen from more than one predetermined region of the genome. 
     
     
         5 . The method of  claim 3 , wherein the DNA fragments are chosen from more than one chromosome. 
     
     
         6 . The method of  claim 1 , wherein the calibration value is determined using a calibration plasma sample including non-tumor DNA and tumor DNA. 
     
     
         7 . The method of  claim 1 , wherein the first parameter represents an abundance of small DNA fragments relative to an abundance of large DNA fragments, and wherein the small DNA fragments have a smaller size than the large DNA fragments. 
     
     
         8 . The method of  claim 1 , further comprising:
 calculating one or more second values of one or more second parameters based on the amounts of DNA fragments at multiple sizes, the one or more second parameters providing different statistical measures of the size profile of DNA fragments in the plasma sample;   comparing the one or more second values to corresponding second calibration values; and   estimating the fractional concentration of tumor DNA in the plasma sample based on the comparisons involving the first value and the one or more second values.   
     
     
         9 . The method of  claim 3 , wherein:
 a first calibration data point specifies a fractional concentration of fetal DNA corresponding to the calibration value of the first parameter;   one or more second calibration data points specify the fractional concentration of tumor DNA corresponding to the one or more second calibration values of the one or more second parameters; and   the first calibration data point and the second calibration data points are points on a multidimensional curve and the comparison includes identifying a multidimensional point having coordinates corresponding to the first value and the one or more second values.   
     
     
         10 . The method of  claim 1 , wherein:
 a calibration data point specifies a fractional concentration of tumor DNA corresponding to the calibration value of the first parameter; and   the calibration data point is determined from a histogram corresponding to a different calibration sample, wherein the histogram provides amounts of DNA fragments at a plurality of sizes, and wherein at least a portion of the different calibration samples have different fractional concentrations.   
     
     
         11 . The method of  claim 1 , wherein the one or more sequence reads comprises a full length sequence of the DNA fragment. 
     
     
         12 . The method of  claim 1 , wherein the one or more sequence reads is obtained from a circularized DNA fragment. 
     
     
         13 . The method of  claim 1 , wherein the one or more sequence reads is obtained from a nanopore-based sequencer. 
     
     
         14 . The method of  claim 1 , wherein the one or more sequence reads is obtained by massively parallel sequencing. 
     
     
         15 . The method of  claim 14 , further comprising performing massively parallel sequencing to obtain the one or more sequence reads for each of the plurality of DNA fragments. 
     
     
         16 . The method of  claim 1 , further comprising:
 determining a function that approximates the calibration value across a plurality of fractional concentrations.   
     
     
         17 . The method of  claim 1 , wherein the organism is a human, and the reference genome is a human reference genome. 
     
     
         18 . The method of  claim 1 , wherein the plurality of DNA fragments comprises greater than 100,000 DNA fragments. 
     
     
         19 . A computer product comprising non-transitory computer readable medium storing a plurality of instructions that when executed control a computer system to analyze a plasma sample of an organism, the plasma sample including cell-free DNA fragments originating from normal cells and potentially from cells associated with cancer, the instructions comprising:
 for each of a plurality of DNA fragments from the plasma sample:
 receiving one or more sequence reads obtained from a sequencing of the DNA fragment, the one or more sequence reads including both ends of the DNA fragment; 
 aligning the one or more sequence reads to a reference genome to obtain aligned locations for both ends of the DNA fragment; and 
 using the aligned locations to determine a size of the DNA fragment; 
   for each size of a plurality of sizes:
 determining an amount of a set of the plurality of DNA fragments from the plasma sample corresponding to the size, using the sizes determined from the aligned locations for the set of the plurality of DNA fragments; 
   calculating a first value of a first parameter based on the amounts of DNA fragments at multiple sizes, the first parameter providing a statistical measure of a size profile of DNA fragments in the plasma sample;   comparing the first value to a calibration value; and   estimating a fractional concentration of tumor DNA in the plasma sample based on the comparison.   
     
     
         20 . A computer system comprising:
 the computer product of  claim 19  and one or more processors.