LEfSe issue at the C) step

[Isabelle Laforest-Lapointe]

Hi,

I'm trying to go through your LEfSe platform on galaxy and I keep getting an error at the C) step.

This is the message I get: No differentially abundant features found in /home/usr/local/galaxy-dist/database/files/000/242/dataset_242547.dat

And it says that it is empty.

I join my data so you can take a look. I tried different things to format it as much as possible as your mouse data.

I'm putting the site as class and species as subclass.

Thanks for your help,

Isabelle Laforest-Lapointe

UQAM

SB-2665

[George Weingart]

Hello Isabelle,

I checked the dataset you mentioned in the server and it is not empty.

Can you send me the initial dataset you are posting to lefse so I check ?

Thanks and best regards,

George Weingart, PhD

Huttenhower Lab

[george.weingart@huttenhower 242]$ wc -l  dataset_242547.dat

165 dataset_242547.dat

[george.weingart@huttenhower 242]$ cat  dataset_242547.dat

Bacteria.Chlorobi.Ignavibacteria        0.0                     -

Bacteria.Bacteroidetes.Cytophagia       4.8107769983                    -

Bacteria.TM7.TM7_3      3.9062666054                    -

Bacteria.TM7.TM7_1      2.05781819443                   -

Bacteria.Chlorobi.SJA_28        1.17084820364                   -

Bacteria.GN02.BB34      0.0                     -

Bacteria.Chloroflexi._Thermobacula_     0.0                     -

Bacteria.Acidobacteria.GAL08    0.0                     -

Bacteria.Nitrospirae.Nitrospira 1.64207628002                   -

Bacteria.Proteobacteria.Gammaproteobacteria     4.93230461361                   -

Bacteria.Acidobacteria.Sva0725  1.87932530078                   -

Bacteria.Actinobacteria.Actinobacteria  4.78220711456                   -

Bacteria.FBP    3.11054930893                   -

Bacteria.Chloroflexi.C0119      2.16208624024   

etc.  etc. etc.

[George Weingart]

Hi Isabelle,

I ran lefse  with the dataset you provided and I get 

"No features with significant differences between the two classes"

Best regards,

George

[jfg]

Hey there Gal.Lab'ers,

   I'm trying to analyse the metabolic characteristics output by PICRUSt in the Galaxy cluster, using HUMANn and LEfSe. When I run my metabolic outputs (attached) through  LEfSe i get no output and a message saying: 

"No differentially abundant features found in /home/usr/local/galaxy-dist/database/files/000/251/dataset_251946.dat"

- just like Isabelle. 

   I'd take this as a sign that there's nothing interesting going on, but I used the sample dataset you give in your tutorial ('129'; body sites and O2), and (as well as having to remove the O2 levels for a 'string to float' error) I get:

"No differentially abundant features found in /home/usr/local/galaxy-dist/database/files/000/251/dataset_251949.dat"

   I understand that having "No differentially abundant features" would produce this error, giving an empty graphical output: but it seems unlikely that my set and the data set you provide would be so lacking in features. Have you any ideas as to where I might be going astray?

   

Workflow:

1. Get Data> Upload File> FileFormat:Tabular File:04-LEfSe.txt / 129.txt

2. Format...> Vectors:Rows Class:1#NAME / #2ID Subclass:none Subject:none Norm:Yes > Execute

3. LEfSe> all settings at default > Execute

4. PLot..> all settings at default > Execute

  -> provides the output as above.

Attached:
- input (metabol from PICRUSt)

- input from tutorial: 129.txt set 

Thanks for the help and time

jfg

[jfg]

PS: could the lack of features have anything to do with the column of hyphens showing on the far right of this dataset screenshot (again, the 129 tutorial data, at the LEfSe stage)?

[jfg]

Problem Resolved: 

   If you encounter this "error", check what your are comparing (e.g. write it out on a piece of paper) and then make sure the class / subclass / subject drop-down boxes in LEfSe> A)Format_Data_for_LEfSe are set to model your relationship of interest properly.