Linear Single-Cell Genome Amplification for Accurate SNV Detection

The patent segments the amplification process into two distinct phases: linear amplification phase where each product is copied only from the original template (not from other products), followed by exponential amplification phase. This segmentation ensures that amplification errors are not propagated during the linear phase, while still achieving sufficient yield through the subsequent exponential phase. The linear amplification uses strand-displacement polymerase with specific primers that prevent product re-priming, creating a controlled environment where errors made in early cycles are not copied by other amplified products.

The patent changes the fundamental parameter of amplification kinetics from exponential to linear by using strand-displacement polymerase with specific primer characteristics (40%-60% G-rich or C-rich sequences with restriction endonuclease sites). This parameter change in the amplification mode eliminates sequence-dependent bias because each template is copied independently at a constant rate, rather than experiencing the competitive dynamics of exponential PCR where certain sequences amplify preferentially. The linear amplification maintains uniform representation across the entire genome.

The patent applies preliminary action by performing linear amplification before sequencing, where amplification errors are prevented from propagating in the first place. During linear amplification, each DNA product is copied only from the original template molecule, not from other amplified products. This preliminary error prevention strategy ensures that amplification errors are independently generated and diluted among linearly amplified products, rather than being propagated and concentrated as in exponential amplification. Consequently, increased sequencing depth can effectively distinguish true variants from errors without inflating false positive rates.