Accuracy of 3% OTUs constructed from denoised pyrosequences from the three multi-sample artificial community datasets retained by APDP and three alternative denoising approaches.

The number of false positives (FP) and false negatives (FN) is shown for cases where miscalled and near-match OTUs are considered to be incorrect (TP only) or correct (TP+NM). In addition, the results are shown for each method after excluding OTUs with fewer reads than the minimum detection threshold of APDP (cutoff).本结果展示了两种判定场景下的假阳性（false positives，FP）与假阴性（false negatives，FN）数量：当错判的（miscalled）与近匹配的（near-match）操作分类单元（Operational Taxonomic Unit，OTU）被认定为错误时（此时仅将真阳性（true positive，TP）计为正确），以及当它们被认定为正确时（此时将真阳性（TP）与近匹配（near-match，NM）均计为正确）。此外，本结果还展示了剔除读数量低于APDP最低检测阈值（截断值cutoff）的OTU后，各分析方法的对应结果。