New in v11
Which taxonomy database should I use?
Naive Bayesian Classifier algorithm
Cross-validation by identity
The nbc_tax command is an implementation of the RDP Naive Bayesian
Classifier algorithm (Wang et al. 2007). Predictions
by nbc_tax are very similar to the Java implementation provided by RDP;
differences are consistent with the randomness of bootstrapping so I believe
that the results are equivalent for all practical purposes..
Taxonomy predictions are written to the -tabbedout file. The first three fields are (1) query sequence label, (2) prediction with
bootstrap values and (3) strand. If the -sintax_cutoff option is given then predictions are written a second time after applying the confidence threshold, keeping only ranks with high enough confidence.
The sintax_summary command can be used
to generate a summary report.
The strand option must be specified.
Multithreading is supported.
usearch -nbc_tax otus.fa -db ref16s.fa -strand plus -tabbedout
et al. (2007) Naive Bayesian classifier for rapid assignment of rRNA
sequences into the new bacterial taxonomy. AEM 73,
References (please cite)
R.C. Edgar (2016), SINTAX: a simple non-Bayesian taxonomy classifier for 16S and ITS sequences, https://doi.org/10.1101/074161
• SINTAX taxonomy prediction algorithm
• Fast and simple method, accuracy comparable to RDP Classifier
R.C. Edgar (2018), Accuracy of taxonomy prediction for 16S rRNA and fungal ITS sequences, PeerJ 6:e4652
• Cross-validation by identity, novel benchmark strategy enabling realistic accuracy estimates
• Genus accuracy of best methods is 50% on V4 sequences
• Recent algorithms do not improve on RDP Classifier or SINTAX
R.C. Edgar (2018), Taxonomy annotation and guide tree errors in 16S rRNA databases, PeerJ 6:e5030
• Approx. one in five SILVA and Greengenes taxonomy annotations are wrong
• SILVA and Greengenes trees have pervasive conflicts with type strain taxonomies