Guardant Health has filed a patent for a method of analyzing DNA samples by partitioning them into hypermethylated and hypomethylated partitions. The method involves tagging the DNA and obtaining sequence reads, grouping them into families, and calling bases based on the sequences. The patent claims that calling certain transition mutations requires observation in a greater number of molecules in the hypermethylated partition compared to the hypomethylated partition. GlobalData’s report on Guardant Health gives a 360-degree view of the company including its patenting strategy. Buy the report here.
According to GlobalData’s company profile on Guardant Health, Personalized medicine biomarkers was a key innovation area identified from patents. Guardant Health's grant share as of September 2023 was 26%. Grant share is based on the ratio of number of grants to total number of patents.
Method for analyzing dna based on methylation patterns
A recently filed patent (Publication Number: US20230313288A1) describes a method for analyzing DNA samples. The method involves partitioning the DNA sample into hypermethylated and hypomethylated partitions and tagging the DNA in these partitions to generate tagged nucleic acids with molecular barcodes. Sequence reads of molecules from both partitions are obtained and grouped into families based on molecular barcode sequences or genomic positions. The method then determines a first set of sequences from the hypermethylated partition and a second set of sequences from the hypomethylated partition. Based on these sets of sequences, the method calls a plurality of bases, specifically C to T or G to A transition mutations, relative to a reference sequence. The calling of these mutations is based on the observation of the mutations in a greater number of molecules from the first set compared to the second set, or the absence of calling the mutations based on the first set and the presence of calling the mutations based on the second set.
The patent also describes variations in the method, such as using different thresholds for calling the mutations based on the first and second sets of sequences. The thresholds provide different levels of specificity for calling the mutations. Additionally, the patent discusses the use of position-specific background error rates for the sequences of molecules from the first and second sets. The calling of the mutations based on the first set requires the observation of the mutations at a frequency exceeding the corresponding rate from the background error rates.
Furthermore, the patent mentions the possibility of obtaining sequence reads from an intermediate partition and determining a third set of sequences. The calling of C to T and G to A transition mutations based on the third set can be less stringent or the same as the calling based on the second set. The DNA samples analyzed in the method can be cell-free DNA or from subjects with or suspected of having a proliferative disorder or solid tumor.
Overall, this patent presents a method for analyzing DNA samples by partitioning them into hypermethylated and hypomethylated partitions, tagging the DNA, obtaining sequence reads, and calling C to T and G to A transition mutations based on different sets of sequences. The method offers variations in thresholds, background error rates, and the inclusion of an intermediate partition for analysis.