Hello everyone! More of a high level anecdotal question. I passed my mouse cecum data through the standard database with taxprofiler and am getting an average of about 45-50% assignment of my reads. Has anyone worked with mouse gut or cecum data that can give me some insight on whether this is a good number or if I should keep trying larger datasets?

I've not analysed that type of data, but I think 45-50% isn't too shabby. This is old (pre 2021) so tkae a pinch of salt! These were my classification statistics for various ancient and modern human related samples using MALT

Thanks so much for finding this! That's awesome! I had done a little literature digging and found a fairly recent custom database made from mining ~1k mouse gut datasets and gave it a whirl with the pipeline. Using this database I achieved ~90% assignment! Granted a lot of the new assignments are not-well-studied organisms, but still impressive so I wanted to share in case it is useful for others! 🙂 Database: https://ezmeta.unige.ch/CMMG/ Publication:https://journals.plos.org/ploscompbiol/article?id=10.1371/journal.pcbi.1009947#sec002

Haha, thank you! I made it myself like 6 years ago during my PhD 🤣

That's funny I didn't even realize! 😄

Shows how important databases are... Yet so few people invest in them :/

Absolutely! I'm impressing that upon the student I'm doing this analysis for!