SRP140497 PRJNA450384 Enhanced Duplex Sequencing methodology for high efficiency error suppression at variable sequencing depths Detection of cancer-associated somatic mutations has broad applications for oncology and precision medicine. However, this becomes challenging when DNA from cancer cells is in low abundance such as in impure tissue specimens or in circulating cell-free DNA, resulting in low allele frequency mutations. Next-generation sequencing (NGS) is particularly prone to technical artefacts that can limit the accuracy for calling low allele frequency mutations. A promising strategy to suppress these technical artefacts utilizes unique molecular identifiers (UMIs) to amalgamate reads from the same DNA template into a consensus sequence. Current UMI-based methods depend on redundant sequencing of template molecules, which leads to poor efficiency and higher sequencing costs. Here, we present a novel and generalizable strategy that maximizes the efficiency of UMI-based methods by utilizing single reads (singletons) for error suppression. By incorporating this “Singleton Correction” approach, we show that up to 23% of singletons not previously utilized for UMI-based error suppression could be retained with error profiles comparable to the gold standard duplex consensus sequences (DCS). Singleton Correction augmented DCS recovery by 4-fold, leading to improved analytical sensitivity while maintaining equivalent specificity. Through downsampling, we demonstrate that by maximizing in silico DCS recovery, Singleton Correction reduced the dependence of DCS recovery on sequencing depth. Additionally, we expanded UMIs with sequence alignment properties to enhance molecular characterization, which resulted in up to a 70% reduction of errors at ultra-deep sequencing. Together, our method combines complex UMIs with retention of high quality singletons to overcome traditional hurdles associated with NGS-based mutational profiling.