The paper introduces Deblur, a novel sub-operational-taxonomic-unit (sOTU) method that addresses the limitations of traditional operational taxonomic unit (OTU) approaches in high-throughput 16S ribosomal RNA gene amplicon sequencing. Deblur uses error profiles to identify putative error-free sequences from Illumina MiSeq and HiSeq platforms, significantly reducing computational demands while maintaining or improving sensitivity and specificity. The method is demonstrated to detect closely related bacterial sequences with single nucleotide differences, removing false positives and maintaining stability in detection. Deblur operates on a per-sample level, making it scalable to large data sets and meta-analyses. The authors compare Deblur with other sOTU methods, such as DADA2 and UNOISE2, showing that Deblur is more stable and efficient, especially for large-scale data sets. Deblur is open-source and freely available, making it a valuable tool for microbiome research.The paper introduces Deblur, a novel sub-operational-taxonomic-unit (sOTU) method that addresses the limitations of traditional operational taxonomic unit (OTU) approaches in high-throughput 16S ribosomal RNA gene amplicon sequencing. Deblur uses error profiles to identify putative error-free sequences from Illumina MiSeq and HiSeq platforms, significantly reducing computational demands while maintaining or improving sensitivity and specificity. The method is demonstrated to detect closely related bacterial sequences with single nucleotide differences, removing false positives and maintaining stability in detection. Deblur operates on a per-sample level, making it scalable to large data sets and meta-analyses. The authors compare Deblur with other sOTU methods, such as DADA2 and UNOISE2, showing that Deblur is more stable and efficient, especially for large-scale data sets. Deblur is open-source and freely available, making it a valuable tool for microbiome research.