Systems and methods to detect rare mutations and copy number variation

What is this patent about?

’172 is related to the field of molecular diagnostics and bioinformatics, specifically focusing on the detection of rare genetic variations and copy number alterations within cell-free polynucleotides. The technology addresses the challenge of identifying low-frequency mutations, such as those shed by tumors into the bloodstream, which are often obscured by the inherent noise and bias of standard sequencing workflows.

The underlying idea behind ’172 is the use of a digital sequencing framework that employs molecular barcodes to track and collapse sequence reads into high-fidelity consensus sequences. By tagging parent molecules before amplification, the system can distinguish between true biological variants and errors introduced during PCR or sequencing, effectively filtering out noise to reveal mutations present at extremely low concentrations.

The claims of ’172 focus on a longitudinal method for assessing cancer by analyzing samples from a subject at multiple time points. The process involves attaching a limited set of barcodes to a population of cell-free DNA (cfDNA) such that the number of unique tags is fewer than the number of molecules mapping to a specific genomic position, followed by amplification and sequencing to monitor genetic changes over time.

In practice, the invention works by grouping sequencing reads into familial sets based on their unique identifiers and genomic coordinates. This familial collapsing allows the system to identify a single parent molecule from many progeny reads, thereby neutralizing amplification bias and providing a quantitative measure of unique molecules. This statistical approach enables the detection of variants with a sensitivity as high as 0.1%.

This method differs from prior approaches by maximizing conversion efficiency and utilizing non-unique tagging strategies to manage high-diversity samples without requiring billions of distinct barcodes. By comparing normalized sequence data across predefined genomic regions and across different time intervals, the system provides a dynamic profile of a subject’s tumor burden, facilitating early detection and therapy monitoring.