I am new to metagenomic analysis using Galaxy and would like to ask about reference databases for taxonomic classification.
Is it possible to upload and use custom reference databases, such as the Ribosomal Database Project (RDP) or Greengenes, in Galaxy for metagenomic analysis? If so, what is the recommended procedure for incorporating these databases into a workflow? Alternatively, does Galaxy already provide these databases, or are there other reference databases that are recommended for metagenomic taxonomic profiling?
Yes, you can use custom reference databases in Galaxy, but the exact process depends on which taxonomic classification tool you’re using. Some tools allow you to upload a pre-built database directly, while others require the database to be formatted or indexed in a specific way before it can be used. That’s usually the step that takes the most time. If you’re just getting started, I’d recommend first using one of the reference databases already supported by your Galaxy server so you can become familiar with the workflow. Once everything is working, it’s much easier to substitute your own database if your project requires it. Also keep in mind that Greengenes hasn’t been updated for quite some time, so many users now prefer more actively maintained reference datasets depending on whether they’re working with 16S amplicons or shotgun metagenomics. The best choice often depends on your sequencing data and the level of taxonomic resolution you need.
As @DavidBaker explained, the public UseGalaxy servers already host many of these indexes. This is a really good question, so I’ll try to summarize a bit more about the Galaxy resources we have to support the kind of work you want to do.
For examples, please see the Galaxy Training Network (GTN) tutorials. These have example paths through common tool choices you can explore!
See the bottom of tool forms for direct links to tutorials that happen to include that tool!
Most tutorials will include a workflow template you can use, but we also have production HTP workflow templates! These are polished versions to use when working with larger batches of data. Workflows ensure the data all process all the same way. They can also help these complex tools to process quicker (an important consideration with this kind of analysis).
Then, for your specific questions, you can do things like:
Load data, organize it into collection folders, run some QA
Then filter with a tool like Filter with SortMeRNA of ribosomal RNAs in metatranscriptomic data.
As shown in the tutorial linked on that tool form:
Try a run with what is already indexed, as @DavidBaker also explained (to get oriented), then you can try loading your own custom index to see what happens.
Warning: the public cluster resources are significant but you’ll need to also be practical here! The UseGalaxy.eu server can sometimes scale the largest, so if a job actually fails for resources somewhere else, and technical issues are eliminated, try there next!
You can explore the tools you plan to use and check for what is hosted at the server where you are working. The UseGalaxy servers Use Galaxy | Galaxy Hub host Kraken2 with all of the same pre-built versions most would be using even outside of Galaxy.
Then, see the bottom of the Kraken2 tool form for example workflows in GTN tutorials.
Finally, you can search the list of databases on the tool form like this. Kraken2 indexes are too large to use as a custom index but if the wanted index is public and in general use, you can request it at this forum and we’ll consider adding it in!
