**3.2 Principles and benefits of dPCR**

The basis of ddPCR is to distribute a nucleic acid-containing sample to thousands of independent partitions. There are several ways to create these droplets: manual partitioning, immiscible liquid chemistry, atomization, etc. The generated droplets contain only one target DNA molecule or none at all [56]. To determine target DNA copies without bias, templates with target DNA must be randomly distributed and microdroplets must be produced in large numbers. These partitions can be individually amplified through thermal cycling. Unlike qPCR, which produces an exponential signal and quantifies samples by comparing their CTs (threshold cycle) to a standard curve generated by well-defined samples; by determining the concentration of the sample using an "analog" method, ddPCR technology generates linear digital signals that allow quantitative analysis of the PCR product, being able to detect very rare mutations with high precision and sensitivity, and these amplicons can be quantified without a curve standard [53]. Quantification of DNA molecules is performed by a combination of Poisson distribution and dilution templates at the single-molecule level. The number of templates correlates positively with positive wells, so the exact number of template copies can be calculated. The use of the presence and absence of signals to indicate the target DNA makes a direct "digital" measurement of the samples [57] (**Figure 4**). Furthermore, due to high pipetting handling in sample preparation and PCR conditions, even with a standard curve, the data disparity in qPCR is greater than in ddPCR. ddPCR can be used to detect low concentrations of DNA [58, 59].
