Lefse analysis in R

Downstream analysis and statistics LEfSe
1

Hi, there

I’m developing a R package for microbiome marker discovery named microbiomeMarker, and the algorithm from lefse has been integrated to this package.

You can try it out today, if you want run lefse analysis in R. microbiomeMarker is still a newborn, so there may be bugs.

Thanks. Any suggestions and contribution will be highly appreciated.

2

Hi -
Thanks for the interests and for the package! If you have any LEfSe related questions, please feel free to let us know. If you are interested in additional microbial biomarker testing methods, the group also has the MaAsLin2 method with functionalities such as random effects for longitudinal designs, etc.
Best,
Siyuan

3

Thanks for sharing~~~
may i ask some questions?

  1. format of otu table should be .txt files? .CSV files? or others?
  2. i get OTU table from “CLC” software
  3. i upload .csv file and run analyze --> show the code" Error: ps must be phyloseq object "
    i do not know how to solve it!! Thanks
4

Hi there,

Thanks for your interest! If you would like to use one of the BioBakery tools we would recommend the use of MaAsLin2(https://github.com/biobakery/maaslin2). If you have questions about the microbiomeMarker R package we would suggest directing your questions to the developer at https://github.com/yiluheihei/microbiomeMarker.

Best!
Kelsey

5

The main input of microbioMarker must be in phyloseq object, which means you must convert your otu table to phyloseq. As Kelsey_Thompson said, if you have questions about the microbiomeMarker R package, you can direct you questions at https://github.com/yiluheihei/microbiomeMarker.

Thanks for you interest.

6

Ying Taur has written a very nice phyloseq object to Lefse R add-on.

An R implementation of lefse is also available from the Waldron lab:

1 Like
7

Unfortunately, there are quite some limitations and if I am not wrong lefser and microbioMarker only supports two-group contrasts.

Note. lefser supports only two-group contrasts.

8

yingtools2 perform lefse analysis by calling the python script. lefser has very limited functionality. microbiomeMarker almost achieved all features of the original python lefse, of course, includs multiple-group constrasts. Moreover, some other common differential analysis methods are also integrated into microbiomeMarker, e.g. deseq2, metagenomeSeq.

1 Like
9

Hi @yiluheihei ,

I want to do the compositional analysis for my data
My study approach is 2 x 2 factorial design i.e 2 genetic lines and 2 efficiencies ( for example, E line high and low, Z line high and low efficiency). I want to see the interaction of 2 genetic lines with the month as a random effect. Is there any way I can do the compositional analysis?
could you suggest to me which package and approach would be appropriate for my study

thanks

10

Hi @yiluheihei,
I am getting one error in microbiomemarker package.

I have used this commands -
mm_lefse ← run_lefse(
ps,
taxa_rank = “Genus”,
norm = “rarefy”,
group = “bodysite”,
subgroup = “genotype”
)

plot_cladogram(mm_lefse, color = c(“brown”,“blue”,“red”,“green”))

Error coming:
Error in if (n <= length(x)) { : missing value where TRUE/FALSE needed

I have also brought this issue in the microbiomeMarker github page.

Please help!

thanks,