taxonomy assignment for de novo OTU picking

Hello,I have obtained some paired end demultiplexed sequence and then imported them to qiime2. After that I ran dada2 to Denise the sample and ran the commands for de novo clustering. Now I am trying to assign taxonomy to the OTU.Every time I try to use the pre trained classifier an error message shows up that wants me to retrain the classifiers.Can you please tell me how to get the newest version of the trained classifier and the.qza file associated with it.
The error is as followed: **Plugin error from feature-classifier:**I can not train my own classifier due to various reasons.

The scikit-learn version (0.22.1) used to generate this artifact does not match the current version of scikit-learn installed (0.24.1). Please retrain your classifier for your current deployment to prevent data-corruption errors.

Another issue that I keep facing is

If you search the forum for does not match the current version of scikit-learn installed, you'll find many threads discussing the solution to this.

In brief, you need to use a classifier from the same version of QIIME 2 that you are currently using. So, if you are using 2022.2, then you need to download the classifiers made in 2022.2.

Hello, thank you for your kind reply. I had went over the post previously but there were tried to follow the solutions but they did not point me to a place where I might be able to won load the newest classifier. The third one that comes up still is giving me the same error.

Figure out what version of Qiime2 you are using, then go to the data resources page, select the version of Qiime2 that matches your version, and download the classifier for that version.

1 Like

This topic was automatically closed 31 days after the last reply. New replies are no longer allowed.