Systems and methods for analyzing circulating tumor DNA
The invention provides oncogenomic methods for detecting tumors by identifying circulating tumor DNA. A patient-specific reference directed acyclic graph (DAG) represents known human genomic sequences and non-tumor DNA from the patient as well as known tumor-associated mutations. Sequence reads from cell-free plasma DNA from the patient are mapped to the patient-specific genomic reference graph. Any of the known tumor-associated mutations found in the reads and any de novo mutations found in the reads are reported as the patient's tumor mutation burden.
1. A method for analyzing tumor DNA, the method comprising:
obtaining a genomic reference graph that represents a plurality of known human genomic sequences, the genomic reference graph comprising a directed graph having vertices and edges stored as objects in a tangible memory subsystem of a computer system, wherein matching homologous portions of the sequences are each represented by a single object and portions that vary are represented as alternate objects, such that each of the plurality of known human genomic sequences is represented as a path in the genomic reference graph;
creating a patient-specific genomic reference graph from the genomic reference graph, the creating comprising: aligning a first set of sequencing reads, obtained by sequencing a sample containing non-tumor DNA from the patient, to the genomic reference graph; identifying mutations relative to the genomic reference graph; and incorporating the identified mutations as alternate objects into the genomic reference graph to create the patient-specific genomic reference graph;
aligning a second set of sequence reads, obtained by sequencing a sample containing cell-free plasma DNA from the patient, to the patient-specific genomic reference graph to find at least one mutation in the cell-free plasma DNA relative to the non-tumor DNA from the patient; and
providing a report that circulating-tumor DNA (ctDNA) in the patient includes the at least one mutation found in the cell-free plasma DNA.
2. The method of claim 1 , wherein the report identifies a patient's tumor-related mutation population, wherein the tumor-related mutation population includes all mutations in the cell-free plasma DNA relative to the non-tumor DNA from the patient.
3. The method of claim 2 , wherein the genomic reference graph is annotated to describe one or more known tumor-associated mutations, and the report further includes what portion of the second set of sequence reads align to mutation-associated branches in the patient-specific genomic reference graph.
4. The method of claim 2 , wherein the report identifies a first clone and a second clone present in a tumor in the patient.
5. The method of claim 2 , wherein the report identifies a driver mutation not present in the known human genomic sequences and present in the cell-free plasma DNA from the patient.
6. The method of claim 5 , further comprising adding the driver mutation as a known tumor-associated mutation in the patient-specific genomic reference graph for subsequent uses.
7. The method of claim 2 , further comprising building the genomic reference graph by:
obtaining the known human genomic sequences;
finding matching homologous portions among the known human genomic sequences;
creating one of the objects in the tangible memory subsystem for each of the matching homologous portions; and
creating connections between pairs of the objects to create the genomic reference graph.
8. The method of claim 7 , further comprising aligning a third set of sequence reads to the patient-specific genomic reference graph and producing a second report that identifies a patient's second tumor-related mutation population at a time different from an initial time associated with the patient's tumor-related mutation population, wherein the second report includes a comparison of the patient's second tumor-related mutation population to the patient's tumor-related mutation population.
9. The method of claim 7 , wherein the objects of the patient-specific genomic reference graph include pointers to adjacent ones of the objects such that the objects are linked into paths to represent the plurality of known human genomic sequences, wherein each pointer identifies a physical location in the memory subsystem at which the adjacent object is stored.
10. The method of claim 9 , further comprising obtaining the second set of sequence reads by sequencing the sample containing the cell-free plasma DNA from the patient.
11. A system for analyzing tumor DNA, the system comprising a processor coupled to a memory subsystem having stored therein:
a genomic reference graph representing a plurality of known human genomic sequences, the genomic reference graph comprising a directed graph having vertices and edges stored as objects, wherein matching homologous portions of the sequences are each represented by a single object and portions that vary are represented as alternate objects, such that each of the plurality of known human genomic sequences is represented as a path in the genomic reference graph, the objects stored in the memory subsystem; and
instructions executable by the processor to cause the system to:
create a patient-specific genomic reference graph from the genomic reference graph, the creating comprising: aligning a first set of sequence reads, obtained by sequencing a sample containing non-tumor DNA from the patient, to the genomic reference graph; identifying mutations relative to the genomic reference graph; and incorporating the identified mutations as alternate objects into the genomic reference graph to create the patient-specific genomic reference graph;
align a second set of sequence reads from cell-free plasma DNA of the patient to the patient-specific genomic reference graph to find at least one mutation in the cell-free plasma DNA relative to the non-tumor DNA from the patient; and
provide a report that circulating-tumor DNA (ctDNA) in the patient includes the at least one mutation found in the cell-free plasma DNA.
12. The system of claim 11 , wherein the report identifies a patient's tumor-related mutation population, wherein the tumor-related mutation population includes all mutations in the cell-free plasma DNA relative to the non-tumor DNA from the patient.
13. The system of claim 12 , wherein the genomic reference graph is annotated to describe one or more known tumor-associated mutations, and the report further includes what portion of the second set of sequence reads align to mutation-associated branches in the patient-specific genomic reference graph.
14. The system of claim 12 , wherein the report identifies a first clone and a second clone present in a tumor in the patient.
15. The system of claim 12 , wherein the report identifies a driver mutation not present in the known human genomic sequences and present in the cell-free plasma DNA from the patient.
16. The system of claim 15 , further operable to add the driver mutation as a known tumor-associated mutation in the patient-specific genomic reference graph for subsequent uses.
17. The system of claim 12 , further operable to build the genomic reference graph by:
obtaining the known human genomic sequences;
finding matching homologous portions among the known human genomic sequences;
creating one of the objects in the tangible memory subsystem for each of the matching homologous portions; and
creating connections between pairs of the objects to create the genomic reference graph.
18. The system of claim 17 , further operable to align a third set of sequence reads to the patient-specific genomic reference graph and produce a second report that identifies a patient's tumor-related mutation population at a time different from an initial time associated with the patient's tumor-related mutation population, wherein the second report includes a comparison of the patient's second tumor-related mutation population to the patient's tumor-related mutation population.
19. The system of claim 17 , wherein the objects of the patient-specific genomic reference graph include pointers to adjacent ones of the objects such that the objects are linked into paths to represent the plurality of known human genomic sequences, wherein each pointer identifies a physical location in the memory subsystem at which the adjacent object is stored.
20. The system of claim 17 , further comprising a nucleic acid sequencing instrument, the system further operable to obtain the second set of sequence reads by sequencing the sample containing the cell-free plasma DNA from the patient using the nucleic acid sequencing instrument.