PICRUSt2 v2.6.0 GTDB

78 views
Skip to first unread message

andrea firrincieli

unread,
Jan 12, 2025, 7:54:44 AMJan 12
to picrust-users
Hi Robyn,

I am testing picrust2 with the GTDB v214 (thank you for this). 

However, after looking at the output files marker_predicted_and_nsti.tsv looks like the names of the reference genomes are still those from the IMG database (see below)

sequence        16S_rRNA_Count  metadata_NSTI   closest_reference_genome
TUSCIA_HTR401_52434014_52434014 1       0.007644        2516653019
TUSCIA_HTR402_52434021_52434021 1       0.013196        2516493030
TUSCIA_HTR404_52434045_52434045 1       0.016127        2516493028
TUSCIA_HTR406_52434069_52434069 1       0.02743 2516493030
TUSCIA_HTR410_52434106_52434106 1       0.012195        2517093017
TUSCIA_HTR411_52434113_52434113 1       0.057822        2648501420
TUSCIA_HTR412_52434120_52434120 1       0.036364        2531839683
TUSCIA_HTR413_52434137_52434137 1       0.027298        2517093029
TUSCIA_HTR417_52434175_52434175 1       0.010767        2654587752

Is this expected, or should we download the new database after the installation?

Also, I can't figure out in which directory of the conda environment the database should be located

Thanks!

ps. For the installation, I have followed the procedure descrived here: here: https://github.com/picrust/picrust2/wiki/Testing-the-updated-PICRUSt2-database

Robyn Wright

unread,
Jan 13, 2025, 10:35:50 AMJan 13
to picrust-users
Hi there,

If you used the same picrust2_pipeline.py command then PICRUSt2 will still run with the previous IMG database (I didn't want to remove compatibility for those that may want to re-run with the previous database for comparability with previous projects), but I have added a new picrust2_pipeline_split.py script that will run with the new database as default. Also at the end of last week, I added some more details on the new database, its performance, and the commands that can be run with it here.

The new database will be in the same folder as the previous database, but where the previous one is inside default_files/prokaryote, the new one is split to default_files/bacteria and default_files/archaea. 

Let me know if you have any other issues!
Thanks,
Robyn
Reply all
Reply to author
Forward
0 new messages