# GTDB classifer building and training **URL:** https://forum.qiime2.org/t/gtdb-classifer-building-and-training/21173 **Category:** User Support **Created:** [October 26, 2021, 3:51pm UTC](https://forum.qiime2.org/t/gtdb-classifer-building-and-training/21173 "2021-10-26T15:51:19Z") **Posts on this page:** 1 **Showing post:** 5
### Author: ![jwdebelius](https://forum.qiime2.org/user_avatar/forum.qiime2.org/jwdebelius/32/9655_2.png) [@jwdebelius](https://forum.qiime2.org/u/jwdebelius) #### Post date: [October 28, 2021, 8:38am UTC](https://forum.qiime2.org/t/gtdb-classifer-building-and-training/21173/5 "2021-10-28T08:38:32Z")
Hi @Kapaiden, Welcome to the :qiime2: forum! I have good news and bad news. The GTDB files at r89 and before are formatted to work with QIIME 2. (There may still be tree issues, but let's burn building the classifier first.) It looks like the later versions don't have that specific set up. I've been working with r89 as a reference database for metagenomic annotation and it works pretty well for my fecal samples, but I am also limited by resources issues. So, I think my first suggestion would be to switch to r89. Once you get the sequences imported, you can follow the [training a feature classifier](https://docs.qiime2.org/2021.8/tutorials/feature-classifier/) tutorial. If you're interested in Silva, you might want to explore the [RESCRIPt](https://forum.qiime2.org/t/processing-filtering-and-evaluating-the-silva-database-and-other-reference-sequence-data-with-rescript/15494) tutorial for Silva specifically. Best, Justine --- _[View the full topic](https://forum.qiime2.org/t/gtdb-classifer-building-and-training/21173)._