The bioBakery help forum

LEfSe reporting a marker as significant when it isn't?

Hi all,

I’ve been using the LEfSe tool to get some candidate enzymatic reactions that are different in my treatment groups. However, it looks like at least one of the hits (with a pretty hefty estimated effect size) isn’t actually significant at all (see EC_3_2_1_23) in the attached figure.

The raw values for this are the following (5 ABT_3wo followed by 3 UT ):
EC_3.2.1.23 23601.13 22159.68 24630.1 25438.25 24401.71 22057.87 22046.91 28232.54

Since the first filter is a rank-sum test for significance, I’m struggling to understand how it came up as a false positive.

Thanks so much for your help.

(upload://1ncQ2qzxXLfkaMtAXHTYlnJqf1P.png)

Hi there,

Just making sure that this doesn’t get completely buried. Any advice on the above would be extremely helpful, thanks so much.

Hi Kevin,

Apologies for my really late response. Can you re-load the png file. I can’t see it. Additionally, without seeing the raw data I am not sure how much help I can be. My guess would be is that the data is not reading-in correctly. Can you re-load the file and we can go forward from there?

Best,
Kelsey