Hello,
I have been told that the instructions at this link (Lefse after QIIME2) creates an input file suitable for testing differences at the genus level and that an input file created where each taxa has a sum is more suitable for testing at all taxonomic levels. Does anyone know how to create this second file in QIIME2?
Thank you,
Akriti
Hello Akiriti,
I was running into this issue as well after following the same steps listed [there] (Lefse after QIIME2) and collapsing the table down to level 6
If I am not mistaken (would love to have a moderator/admin/power user verify or correct my understanding), the problem for us stemmed from when representatives sequences were not able to be resolved with enough specificity, it would simply group those rep seqs, determine the relative abundance, and label those to the most specific one possible (next taxonomic level up), and create a new row with that label. Which for us, was not an issue for the most part as we interpreted it the same way with our external analysis.
For example:
1 Bacteria|Actinobacteria|Actinobacteria|Corynebacteriales|Corynebacteriaceae|Corynebacterium
2 Bacteria|Actinobacteria|Actinobacteria|Corynebacteriales|Corynebacteriaceae|Corynebacterium 1
3 Bacteria|Actinobacteria|Actinobacteria|Corynebacteriales|Corynebacteriaceae|Lawsonella
4 Bacteria|Actinobacteria|Actinobacteria|Corynebacteriales|Corynebacteriaceae
In this case, the row ending in “|Corynebacteriaceae” (Row 4) would only include the relative abundances of those rep seqs that could not be resolved past the Corynebacteriaceae family and does not include Rows 1-3. But the issue is that LEfSe does not interpret those rows in the same manner - instead, it assumes that the “|Corynebacteriaceae” row (Row 4) contains the relative abundance values of the entire Corynebacteriaceae family (including Rows 1-3) regardless of further specificity. This is where we ran into issues.
But what we also noticed was that when taxa were resolved to the same levels or at least one level more specific than the target comparison taxa level, there were not any issues. So using the example above, even though LEfSe was misinterpreting Row 4 as total cumulative relative abundances for the Corynebacteriaceae family, LEfSe did not have issues identifying which relative abundances were corresponding to the Corynebacteriales order as we did not have unresolved rep seqs at that level and thus no row that ended with “|Corynebacteriales”. So we assumed it was able to extrapolate that all relative abundance rows that contained “|Corynebacteriales|” (Rows 1-4 in this example) belonged to the Corynebacteriales order and it ran those values through the LEfSe analysis pipeline correctly.
Thus, our proposed solution was to simply add an additional taxa level label of “|Unknown” to all rows that were not resolved to the L6-Genus level. That way there were no rows that LEfSe was interpreting as cumulative relative abundances for a specific taxa level and instead would be forced to determine that value itself. Additionally, “|Unknown” would not be detected unless those ungrouped rep seqs had a significant enough relative abundance to pass the threshold to begin with.
For example:
1 Bacteria|Actinobacteria|Actinobacteria|Corynebacteriales|Corynebacteriaceae|Corynebacterium
2 Bacteria|Actinobacteria|Actinobacteria|Corynebacteriales|Corynebacteriaceae|Corynebacterium 1
3 Bacteria|Actinobacteria|Actinobacteria|Corynebacteriales|Corynebacteriaceae|Lawsonella
4 Bacteria|Actinobacteria|Actinobacteria|Corynebacteriales|Corynebacteriaceae|Unknown
By doing this, we increased the number of discriminate features detected by LEfSe above our LDA threshold from 11 to 14 for our dataset.
Again, this seemed to have solved our LEfSe issues for testing differences at all levels 1-6 but would like to run this by the community to see if this solution is logical and would not have significant issues.
Thank you,
Daniel
Hi @dann818,
Recently I too have been playing around with the LEfSe + QIIME 2 combination, so I was very happy to come across your post.
I'm not sure I understood 100% of what you wrote, but I generally agree with you that preparing an input file for LEfSe can be tricky, especially with the distinction between relative abundance vs. accumulative relative abundance. For example, when we look at this example input file provided by the original authors (lefse · biobakery/biobakery Wiki · GitHub), each column -- which corresponds to a sample -- doesn't sum to 1 because, for example, the relative abundance of the taxon [ Bacteria ] is the accumulative relative abundance of [ Bacteria|Acidobacteria ], [ Bacteria|Bacteroidetes ], ... Therefore, this is not the same as the QIIME 2 feature table with FeatureTable[RelativeFrequency] type whose column does sum up to 1. And I think this is what you are trying to point out in your post (correct me if I am wrong).
The good news is: I believe LEfSe is smart enough to calculate accumulative relative abundance on its own when it's given a FeatureTable[RelativeFrequency] and it doesn't see accumulative relative abundance. I have written a short tutorial for performing LEfSe with a QIIME 2 feature table. I will copy and paste it below. Take a look at the input_table.tsv file vs. the formatted_table.tsv file. The first file has only 220 taxa with all of them belonging to the genus level; it's the input file for LEfSe. The second file is what LEfSe creates after formatting the first file and it has 429 taxa with varying taxonomic ranks from Kingdom to Genus.
I hope you find my tutorial helpful and please let me know if you have any questions.
In this section, I will walk you through how I run the LEfSe (linear discriminant analysis effect size) tool. But before I do that, it is important for you to acknowledge this:
LEfSe method is more a discriminant analysis method rather than a DA method. (Lin and Peddada, 2020; PMID: 33268781)
In order to use LEfSe, you will need to open two Terminal windows: one for your usual QIIME 2 environment and another for running LEfSe. For the latter, you should create a new conda environment and install LEfSe as described below.
$ conda activate qiime2-2020.8
$ conda create -n lefse -c conda-forge python=2.7.15
$ conda activate lefse
$ conda install -c bioconda -c conda-forge lefse
After you have both terminals set up, you can create an input file for LEfSe from a QIIME 2 feature table. We will use the "Moving Pictures" tutorial as an example (run below in the QIIME 2 terminal).
$ dokdo prepare-lefse \
-t data/moving-pictures-tutorial/table.qza \
-x data/moving-pictures-tutorial/taxonomy.qza \
-m data/moving-pictures-tutorial/sample-metadata.tsv \
-o output/Useful-Information/input_table.tsv \
-c body-site \
-u subject \
-w "[body-site] IN ('tongue', 'gut', 'left palm')"
Click here to view the input_table.tsv file.
Next, we need to format the input table (run below in the LEfSe terminal):
$ lefse-format_input.py \
output/Useful-Information/input_table.tsv \
output/Useful-Information/formatted_table.in \
-c 1 \
-u 2 \
-o 1000000 \
--output_table output/Useful-Information/formatted_table.tsv
Click here to view the formatted_table.in file. Click here to view the formatted_table.tsv file.
We can run LEfSe with (run below in the LEfSe terminal):
$ run_lefse.py \
output/Useful-Information/formatted_table.in \
output/Useful-Information/output.res
Which will give:
Number of significantly discriminative features: 199 ( 199 ) before internal wilcoxon
Number of discriminative features with abs LDA score > 2.0 : 199
Click here to view the output.res file.
We can then list the discriminative features and their LDA scores (run below in the LEfSe terminal):
$ lefse-plot_res.py \
output/Useful-Information/output.res \
output/Useful-Information/output.pdf \
--format pdf
Click here to view the output.pdf file.
Finally, you can create a cladogram for the discriminative features (run below in the LEfSe terminal):
$ lefse-plot_cladogram.py \
output/Useful-Information/output.res \
output/Useful-Information/output.cladogram.pdf \
--format pdf
Click here to view the output.cladogram.pdf file.
Hello @sbslee !!
What a nice code and tutorial! It may will be really helpful for all of us that are facing some problems with this LefSe and Qiime2 combination 
I have tried and followed your tutorial and have reached (almost with no problems, except for some commands I was missing, as Dokdo, for example, which It was not downloaded, neither installed ) to the step where the input table has to be formatted.
lefse-format_input.py \
output/Useful-Information/input_table.tsv \
output/Useful-Information/formatted_table.in \
-c 1 \
-u 2 \
-o 1000000 \
--output_table output/Useful-Information/formatted_table.tsv
I have a couple of questions in this step:
lefse-format_input.py: command not found
It doesn't come to my mind any solution or idea from where I can get or installed this command.
formatted_table.in. What is it? What does it contain? And... is it an input file or an output?**Maybe when I move forward the tutorial I will have more doubts... But those two questions are what I get until now.
Thank you so much in advanced!
I'm looking forward to getting the whole code up 
Bests,
Miriam
Hi @MiriamGorostidi,
Did you install LEfSe in a fresh conda environment as instructed in the tutorial? If you did, before you ran the command, did you activate the conda environment where LEfSe is installed (note that this is not the QIIME 2 environment)? I'm asking these questions because the error suggests your environment can't find LEfSe completely.
The formatted_table.in file is the input file for the command run_lefse.py. It contains relative abundance and accumulative relative abundance data. It is generated from the lefse-format_input.py command and serves as the input file for run_lefse.py. See my tutorial above for more details.
I hope this helps! Let me know if you have additional questions.
Hi @sbslee !
How are you? Thank you so much for your rapid answer!
You were right... I thought I had installed everything, but it seems I did not. I have reached to, almost, the last step. However, I get the next error when creating the last PDF documents:
lefse-plot_res.py
output/Useful-Information/output.res
output/Useful-Information/output.pdf
--format pdflefse-plot_cladogram.py
output/Useful-Information/output.res
output/Useful-Information/output.cladogram.pdf
--format pdf
When running the command above, I get an error that, apparently is attributed to:
clade_sep parameter too large, lowered to 0.266967773438
Traceback (most recent call last):
File "/home/unidad/anaconda3/envs/lefse/bin/lefse-plot_cladogram.py", line 341, in <module>
draw_tree(params['output_file'],clad_tree,params)
File "/home/unidad/anaconda3/envs/lefse/bin/lefse-plot_cladogram.py", line 294, in draw_tree
ax = fig.add_subplot(111, polar=True, frame_on=False, axis_bgcolor=params['back_color'] )
File "/home/unidad/anaconda3/envs/lefse/lib/python2.7/site-packages/matplotlib/figure.py", line 1257, in add_subplot
a = subplot_class_factory(projection_class)(self, *args, **kwargs)
File "/home/unidad/anaconda3/envs/lefse/lib/python2.7/site-packages/matplotlib/axes/_subplots.py", line 77, in __init__
self._axes_class.__init__(self, fig, self.figbox, **kwargs)
File "/home/unidad/anaconda3/envs/lefse/lib/python2.7/site-packages/matplotlib/projections/polar.py", line 854, in __init__
Axes.__init__(self, *args, **kwargs)
File "/home/unidad/anaconda3/envs/lefse/lib/python2.7/site-packages/matplotlib/axes/_base.py", line 541, in __init__
self.update(kwargs)
File "/home/unidad/anaconda3/envs/lefse/lib/python2.7/site-packages/matplotlib/artist.py", line 888, in update
for k, v in props.items()]
File "/home/unidad/anaconda3/envs/lefse/lib/python2.7/site-packages/matplotlib/artist.py", line 881, in _update_property
raise AttributeError('Unknown property %s' % k)
AttributeError: Unknown property axis_bgcolor
Do you know how could I fixed this?
Thank you so much for your help
First of all, it seems like you were able to run lefse-plot_res.py without any issues at least and the error is coming from lefse-plot_cladogram.py. Can you confirm this? For example, can you show me the output/Useful-Information/output.pdf file?
Secondly, you may have one or more outdated libraries installed in your LEfSe environment. Can you show me the result of conda list after activating your LEfSe environment?
(lefse) sbslee@x86_64-apple-darwin13 ~ % conda list
# packages in environment at /Users/sbslee/opt/anaconda3/envs/lefse:
#
# Name Version Build Channel
_r-mutex 1.0.1 anacondar_1 conda-forge
backports 1.0 py_2 conda-forge
backports.functools_lru_cache 1.6.1 py_0 conda-forge
backports_abc 0.5 py_1 conda-forge
biom-format 2.1.7 py27_0 bioconda
bwidget 1.9.14 h694c41f_0 conda-forge
bzip2 1.0.8 hc929b4f_4 conda-forge
ca-certificates 2020.12.5 h033912b_0 conda-forge
cairo 1.16.0 h0ab9d94_1001 conda-forge
cctools_osx-64 949.0.1 h2f0f38f_19 conda-forge
certifi 2019.11.28 py27h8c360ce_1 conda-forge
clang 11.0.1 h694c41f_1 conda-forge
clang-11 11.0.1 default_hf8bb9ca_1 conda-forge
clang_osx-64 11.0.1 hb91bd55_0 conda-forge
clangxx 11.0.1 default_hf8bb9ca_1 conda-forge
clangxx_osx-64 11.0.1 h7e1b574_0 conda-forge
click 7.1.2 pyh9f0ad1d_0 conda-forge
compiler-rt 11.0.1 h654b07c_0 conda-forge
compiler-rt_osx-64 11.0.1 h8c5fa43_0 conda-forge
curl 7.68.0 h8754def_0 conda-forge
cycler 0.10.0 py_2 conda-forge
fontconfig 2.13.1 h1027ab8_1000 conda-forge
freetype 2.10.4 h4cff582_1 conda-forge
fribidi 1.0.10 hbcb3906_0 conda-forge
functools32 3.2.3.2 py_3 conda-forge
future 0.18.2 py27h8c360ce_1 conda-forge
futures 3.3.0 py27h8c360ce_1 conda-forge
gettext 0.19.8.1 haf92f58_1004 conda-forge
gfortran_osx-64 4.8.5 h22b1bf0_8 conda-forge
glib 2.66.3 h519c658_0 conda-forge
graphite2 1.3.13 h2e338ed_1001 conda-forge
gsl 2.4 ha2d443c_1006 conda-forge
h5py 2.10.0 nompi_py27h106b333_102 conda-forge
harfbuzz 2.4.0 h92b87b8_1 conda-forge
hdf5 1.10.5 nompi_h0cbb7df_1103 conda-forge
icu 58.2 h0a44026_1000 conda-forge
jpeg 9d hbcb3906_0 conda-forge
kiwisolver 1.1.0 py27h5cd23e5_1 conda-forge
krb5 1.16.4 h1752a42_0 conda-forge
ld64_osx-64 530 hea264c1_19 conda-forge
ldid 2.1.2 h7660a38_2 conda-forge
lefse 1.0.8.post1 py27_2 bioconda
libblas 3.8.0 14_openblas conda-forge
libcblas 3.8.0 14_openblas conda-forge
libclang-cpp11 11.0.1 default_hf8bb9ca_1 conda-forge
libcurl 7.68.0 h709d2b2_0 conda-forge
libcxx 11.0.1 habf9029_0 conda-forge
libedit 3.1.20191231 h0678c8f_2 conda-forge
libffi 3.2.1 hb1e8313_1007 conda-forge
libgfortran 3.0.1 0 conda-forge
libglib 2.66.3 h2575888_0 conda-forge
libiconv 1.16 haf1e3a3_0 conda-forge
liblapack 3.8.0 14_openblas conda-forge
libllvm11 11.0.1 h223d4b2_0 conda-forge
libopenblas 0.3.7 hd44dcd8_1 conda-forge
libpng 1.6.37 h7cec526_2 conda-forge
libssh2 1.9.0 h8a08a2b_5 conda-forge
libtiff 4.2.0 h355d032_0 conda-forge
libwebp-base 1.1.0 hbcb3906_3 conda-forge
libxml2 2.9.9 hd80cff7_2 conda-forge
linecache2 1.0.0 py_1 conda-forge
llvm-openmp 11.0.1 h7c73e74_0 conda-forge
llvm-tools 11.0.1 h223d4b2_0 conda-forge
lz4-c 1.9.3 h046ec9c_0 conda-forge
make 4.3 h22f3db7_1 conda-forge
matplotlib 2.1.2 py27_1 conda-forge
matplotlib-base 2.1.2 py27h31f9439_1 conda-forge
ncurses 6.2 h2e338ed_4 conda-forge
numpy 1.16.5 py27hde6bac1_0 conda-forge
openssl 1.1.1i h35c211d_0 conda-forge
pandas 0.24.2 py27h86efe34_0 conda-forge
pango 1.42.4 haa940fe_4 conda-forge
pcre 8.44 hb1e8313_0 conda-forge
pip 20.1.1 pyh9f0ad1d_0 conda-forge
pixman 0.38.0 h01d97ff_1003 conda-forge
pyparsing 2.4.7 pyh9f0ad1d_0 conda-forge
pyqi 0.3.2 py27_1 bioconda
python 2.7.15 h8e446fc_1011_cpython conda-forge
python-dateutil 2.8.1 py_0 conda-forge
python_abi 2.7 1_cp27m conda-forge
pytz 2020.1 pyh9f0ad1d_0 conda-forge
r-base 3.5.3 hb1347aa_0
r-codetools 0.2_16 r35h6115d3f_1001 conda-forge
r-coin 1.3_1 r35h159158b_0 conda-forge
r-lattice 0.20_41 r35h17f1fa6_1 conda-forge
r-libcoin 1.0_5 r35h26f5615_1 conda-forge
r-mass 7.3_51.6 r35h17f1fa6_1 conda-forge
r-matrix 1.2_18 r35h26f5615_2 conda-forge
r-matrixstats 0.56.0 r35h17f1fa6_0 conda-forge
r-modeltools 0.2_23 r35h6115d3f_0 conda-forge
r-multcomp 1.4_13 r35h6115d3f_0 conda-forge
r-mvtnorm 1.0_11 r35haf69682_2 conda-forge
r-sandwich 2.5_1 r35h6115d3f_1 conda-forge
r-survival 3.1_12 r35h17f1fa6_0 conda-forge
r-th.data 1.0_10 r35h6115d3f_1 conda-forge
r-zoo 1.8_7 r35h17f1fa6_0 conda-forge
readline 8.0 h0678c8f_2 conda-forge
rpy2 2.8.6 py27r35hfc83f80_2 conda-forge
scipy 1.2.1 py27hab3da7d_2 conda-forge
setuptools 44.0.0 py27_0 conda-forge
singledispatch 3.4.0.3 pyh9f0ad1d_1001 conda-forge
six 1.15.0 pyh9f0ad1d_0 conda-forge
sqlite 3.34.0 h17101e1_0 conda-forge
subprocess32 3.5.4 py27h0b31af3_0 conda-forge
tapi 1100.0.11 h9ce4665_0 conda-forge
tk 8.6.10 h0419947_1 conda-forge
tktable 2.10 h49f0cf7_3 conda-forge
tornado 5.1.1 py27h1de35cc_1000 conda-forge
traceback2 1.4.0 py27_0 conda-forge
unittest2 1.1.0 py_0 conda-forge
wheel 0.36.2 pyhd3deb0d_0 conda-forge
xz 5.2.5 haf1e3a3_1 conda-forge
zlib 1.2.11 h7795811_1010 conda-forge
zstd 1.4.8 hf387650_1 conda-forge
(lefse) sbslee@x86_64-apple-darwin13 ~ %
Finally, can you make sure you are using the exact input files in the tutorial?
Hello @sbslee !
I'm sorry I didn't explain myself so well... I was not able to run lefse-plot_res.py . Instead, I got the following error, so I can't show you the output.pdf file, due to this not being generated.
Traceback (most recent call last):
File "/home/unidad/anaconda3/envs/lefse/bin/lefse-plot_res.py", line 177, in
else: plot_histo_hor(params['output_file'],params,data,len(data['cls']) == 2,params['report_features'])
File "/home/unidad/anaconda3/envs/lefse/bin/lefse-plot_res.py", line 70, in plot_histo_hor
ax = fig.add_subplot(111,frame_on=False,axis_bgcolor=params['back_color'])
File "/home/unidad/anaconda3/envs/lefse/lib/python2.7/site-packages/matplotlib/figure.py", line 1257, in add_subplot
a = subplot_class_factory(projection_class)(self, *args, **kwargs)
File "/home/unidad/anaconda3/envs/lefse/lib/python2.7/site-packages/matplotlib/axes/_subplots.py", line 77, in init
self._axes_class.init(self, fig, self.figbox, **kwargs)
File "/home/unidad/anaconda3/envs/lefse/lib/python2.7/site-packages/matplotlib/axes/_base.py", line 541, in init
self.update(kwargs)
File "/home/unidad/anaconda3/envs/lefse/lib/python2.7/site-packages/matplotlib/artist.py", line 888, in update
for k, v in props.items()]
File "/home/unidad/anaconda3/envs/lefse/lib/python2.7/site-packages/matplotlib/artist.py", line 881, in _update_property
raise AttributeError('Unknown property %s' % k)
AttributeError: Unknown property axis_bgcolor
Here is the output of conda list command. I hope the error is on account of outdated libraries. However, I don't understand how is that possible. I mean, If it is the first time I use LefSe and I have just installed it, how are the libraries not updated? **I'm quite new on command line programming 
conda list # packages in environment at /home/unidad/anaconda3/envs/lefse: # # Name Version Build Channel _libgcc_mutex 0.1 conda_forge conda-forge _openmp_mutex 4.5 1_gnu conda-forge _r-mutex 1.0.1 anacondar_1 conda-forge backports 1.0 py_2 conda-forge backports.functools_lru_cache 1.6.1 py_0 conda-forge backports_abc 0.5 py_1 conda-forge binutils_impl_linux-64 2.35.1 h193b22a_2 conda-forge binutils_linux-64 2.35 hc3fd857_29 conda-forge biom-format 2.1.7 py27_0 bioconda bwidget 1.9.14 ha770c72_0 conda-forge bzip2 1.0.8 h7f98852_4 conda-forge ca-certificates 2020.12.5 ha878542_0 conda-forge cairo 1.16.0 hcf35c78_1003 conda-forge certifi 2019.11.28 py27h8c360ce_1 conda-forge click 7.1.2 pyh9f0ad1d_0 conda-forge curl 7.71.1 he644dc0_3 conda-forge cycler 0.10.0 py_2 conda-forge dbus 1.13.6 he372182_0 conda-forge expat 2.2.10 h9c3ff4c_0 conda-forge fontconfig 2.13.1 hba837de_1004 conda-forge freetype 2.10.4 h0708190_1 conda-forge fribidi 1.0.10 h36c2ea0_0 conda-forge functools32 3.2.3.2 py_3 conda-forge future 0.18.2 py27h8c360ce_1 conda-forge futures 3.3.0 py27h8c360ce_1 conda-forge gcc_impl_linux-64 7.5.0 hda68d29_13 conda-forge gcc_linux-64 7.5.0 he2a3fca_29 conda-forge gettext 0.19.8.1 hf34092f_1004 conda-forge gfortran_impl_linux-64 7.5.0 h56cb351_18 conda-forge gfortran_linux-64 7.5.0 ha081f1e_29 conda-forge glib 2.66.1 h680cd38_0 conda-forge graphite2 1.3.13 h58526e2_1001 conda-forge gsl 2.6 he838d99_2 conda-forge gst-plugins-base 1.14.5 h0935bb2_2 conda-forge gstreamer 1.14.5 h36ae1b5_2 conda-forge gxx_impl_linux-64 7.5.0 h64c220c_13 conda-forge gxx_linux-64 7.5.0 h547f3ba_29 conda-forge h5py 2.10.0 nompi_py27h513d04c_102 conda-forge harfbuzz 2.4.0 h9f30f68_3 conda-forge hdf5 1.10.5 nompi_h7c3c948_1111 conda-forge icu 64.2 he1b5a44_1 conda-forge jpeg 9d h36c2ea0_0 conda-forge kernel-headers_linux-64 2.6.32 h77966d4_13 conda-forge kiwisolver 1.1.0 py27h9e3301b_1 conda-forge krb5 1.17.2 h926e7f8_0 conda-forge ld_impl_linux-64 2.35.1 hea4e1c9_2 conda-forge lefse 1.0.8.post1 py27_1 bioconda libblas 3.9.0 8_openblas conda-forge libcblas 3.9.0 8_openblas conda-forge libcurl 7.71.1 hcdd3856_3 conda-forge libedit 3.1.20191231 he28a2e2_2 conda-forge libffi 3.2.1 he1b5a44_1007 conda-forge libgcc-ng 9.3.0 h2828fa1_18 conda-forge libgfortran-ng 7.5.0 h14aa051_18 conda-forge libgfortran4 7.5.0 h14aa051_18 conda-forge libgomp 9.3.0 h2828fa1_18 conda-forge libiconv 1.16 h516909a_0 conda-forge liblapack 3.9.0 8_openblas conda-forge libopenblas 0.3.12 pthreads_hb3c22a3_1 conda-forge libpng 1.6.37 h21135ba_2 conda-forge libssh2 1.9.0 hab1572f_5 conda-forge libstdcxx-ng 9.3.0 h6de172a_18 conda-forge libtiff 4.2.0 hdc55705_0 conda-forge libuuid 2.32.1 h7f98852_1000 conda-forge libwebp-base 1.2.0 h7f98852_0 conda-forge libxcb 1.13 h7f98852_1003 conda-forge libxml2 2.9.10 hee79883_0 conda-forge linecache2 1.0.0 py_1 conda-forge lz4-c 1.9.3 h9c3ff4c_0 conda-forge make 4.3 hd18ef5c_1 conda-forge matplotlib 2.2.5 ha770c72_3 conda-forge matplotlib-base 2.2.5 py27h250f245_1 conda-forge ncurses 6.2 h58526e2_4 conda-forge numpy 1.16.5 py27h95a1406_0 conda-forge openssl 1.1.1i h7f98852_0 conda-forge pandas 0.24.2 py27hb3f55d8_0 conda-forge pango 1.42.4 h7062337_4 conda-forge pcre 8.44 he1b5a44_0 conda-forge pip 20.1.1 pyh9f0ad1d_0 conda-forge pixman 0.38.0 h516909a_1003 conda-forge pthread-stubs 0.4 h36c2ea0_1001 conda-forge pyparsing 2.4.7 pyh9f0ad1d_0 conda-forge pyqi 0.3.2 py27_1 bioconda pyqt 5.9.2 py27hcca6a23_4 conda-forge python 2.7.15 h5a48372_1011_cpython conda-forge python-dateutil 2.8.1 py_0 conda-forge python_abi 2.7 1_cp27mu conda-forge pytz 2020.1 pyh9f0ad1d_0 conda-forge qt 5.9.7 h0c104cb_3 conda-forge r-base 3.6.3 h316533a_2 conda-forge r-codetools 0.2_18 r36hc72bb7e_0 conda-forge r-coin 1.4_1 r36hcfec24a_0 conda-forge r-lattice 0.20_41 r36hcfec24a_3 conda-forge r-libcoin 1.0_8 r36he454529_0 conda-forge r-mass 7.3_53 r36hcfec24a_0 conda-forge r-matrix 1.3_2 r36he454529_0 conda-forge r-matrixstats 0.58.0 r36hcfec24a_0 conda-forge r-modeltools 0.2_23 r36h6115d3f_1 conda-forge r-multcomp 1.4_16 r36hc72bb7e_0 conda-forge r-mvtnorm 1.1_1 r36h31ca83e_1 conda-forge r-sandwich 3.0_0 r36h142f84f_0 conda-forge r-survival 3.2_7 r36hcfec24a_0 conda-forge r-th.data 1.0_10 r36h6115d3f_2 conda-forge r-zoo 1.8_8 r36hcdcec82_0 conda-forge readline 8.0 he28a2e2_2 conda-forge rpy2 2.8.6 py27r36hd767a1f_2 conda-forge scipy 1.2.1 py27h921218d_2 conda-forge sed 4.8 he412f7d_0 conda-forge setuptools 44.0.0 py27_0 conda-forge singledispatch 3.4.0.3 pyh9f0ad1d_1001 conda-forge sip 4.19.8 py27hf484d3e_1000 conda-forge six 1.15.0 pyh9f0ad1d_0 conda-forge sqlite 3.34.0 h74cdb3f_0 conda-forge subprocess32 3.5.4 py27h516909a_0 conda-forge sysroot_linux-64 2.12 h77966d4_13 conda-forge tk 8.6.10 h21135ba_1 conda-forge tktable 2.10 hb7b940f_3 conda-forge tornado 5.1.1 py27h14c3975_1000 conda-forge traceback2 1.4.0 py27_0 conda-forge unittest2 1.1.0 py_0 conda-forge wheel 0.36.2 pyhd3deb0d_0 conda-forge xorg-kbproto 1.0.7 h7f98852_1002 conda-forge xorg-libice 1.0.10 h516909a_0 conda-forge xorg-libsm 1.2.3 h84519dc_1000 conda-forge xorg-libx11 1.6.12 h516909a_0 conda-forge xorg-libxau 1.0.9 h7f98852_0 conda-forge xorg-libxdmcp 1.1.3 h7f98852_0 conda-forge xorg-libxext 1.3.4 h516909a_0 conda-forge xorg-libxrender 0.9.10 h516909a_1002 conda-forge xorg-renderproto 0.11.1 h14c3975_1002 conda-forge xorg-xextproto 7.3.0 h7f98852_1002 conda-forge xorg-xproto 7.0.31 h7f98852_1007 conda-forge xz 5.2.5 h516909a_1 conda-forge zlib 1.2.11 h516909a_1010 conda-forge zstd 1.4.8 ha95c52a_1 conda-forge
Thank you so much in advanced
Good question! When you installed LEfSe in your conda environment, you actually installed additional packages with specific versions that you see with $ conda list. And these specific versions may be not most up to date. In fact, I think I found the issue. If you compare your list to mine, you will see that I'm using matplotlib 2.1.2 while you are using matplotlib 2.2.5. I'm not exactly sure how you got that particular version in the first place, but it seems like the error you reported (AttributeError: Unknown property axis_bgcolor) is caused because axis_bgcolor was removed since matplotlib 2.2.0 (see the changelog here). To fix the issue, you have to remove matplotlib from your LEfSe environment and then reinstall a version below 2.2.0:
$ conda activate lefse
$ conda remove matplotlib
$ conda install matplotlib=2.1.2
Hope this fixes the problem. Good luck!
Thank you so much for everything you have done! The problem is fixed and I finally got the two PDF files 
Now I have an additional (and I hope final) question: How are we supposed to interpret the results when we get a __ instead of a bacterial name? I have had a look to your PDF-s and you also get that kind of results (comparing with my PDF-s yours is much more completed and longer; I have a really small number of samples). I supposed those are sequences that did not get to be mapped or classified with any bacteria, aren’t them? Thus, how to interpret that? Just ignoring them?
I attached my PDF-s below, so you can have a look at them 
Thank you so much again!
Hi @MiriamGorostidi,
That's an excellent question! For general explanation on how to interpret things like __ and g__, I will link this previous, but clear answer from the forum: Follow-up on "'unique' taxonomy strings that seem to be shared" - #2 by Nicholas_Bokulich
The distinction is that the first row (ending in ; ) cannot be confidently classified beyond family level (probably because a close match does not exist in the reference database). So sequences receiving that classification can be any taxon in f__Geodermatophilaceae. The second row (ending in g__;s__) DOES have a close match in the reference database and hence is confidently classified at species level — unfortunately, that close match does not have genus or species-level annotations. This does not in any way imply that these two different taxonomic affiliations are related beyond the family level, so it would probably be inappropriate (or at least presumptuous) to collapse these at species level.
Now that we are (hopefully) on the same page with what those double underscores mean, what I ended up doing was updating my dokdo package to handle the double underscores better. For example, instead of outputting k__Bacteria|p__Actinobacteria|c__Actinobacteria|o__Actinomycetales|f__Intrasporangiaceae|f__Intrasporangiaceae|__, the 1.6.0-dev version now outputs k__Bacteria|p__Actinobacteria|c__Actinobacteria|o__Actinomycetales|f__Intrasporangiaceae|f__Intrasporangiaceae_x__L6. This way, if this particular taxon is returned as significant by LEfSe, it will show up as f__Intrasporangiaceae_x__L6 instead of __. For more examples, scroll up and click the files linked to my original tutorial; I updated those files. I will link one PDF file here as an example. output.pdf - Google Drive
Finally, if you wish to re-run your analysis with the 1.6.0-dev version of Dokdo, make sure to re-install Dokdo:
$ pip uninstall dokdo
$ git clone https://github.com/sbslee/dokdo
$ cd dokdo
$ git checkout 1.6.0-dev
$ pip install .
Hope this helps and let me know if you have any questions!
Thank you @sbslee !!
Really helpful and clear to understand
Hi @sbslee
I am trying to use LEfSe following the commands you provided here
during preparing the files I got this error (at the end in bold)
dokdo prepare-lefse
-t table.qza
-x taxonomy.qza
-m metadata.tsv
-o output/Useful-Information/input_table.tsv
-c health-state
-u age
-w "[health-state] IN ('healthy', 'patient')"
Traceback (most recent call last):
File "/usr/local/bin/dokdo", line 5, in
from dokdo.main import main
File "/usr/local/lib/python3.8/dist-packages/dokdo/init.py", line 1, in
from .api import *
File "/usr/local/lib/python3.8/dist-packages/dokdo/api/init.py", line 1, in
from .get_mf import get_mf
File "/usr/local/lib/python3.8/dist-packages/dokdo/api/get_mf.py", line 1, in
from qiime2 import Metadata
ModuleNotFoundError: No module named 'qiime2'
I am using qiime2. 2021.4 and python 3.8
any advices?
thank you
It looks like you ran the command ($ dokdo prepare-lefse ...) in an environment where QIIME 2 is not installed. Did you make sure to activate your qiime2-2021.4 environment with $ conda activate before running the command?
thank you for prompt reply
Yes, sure that qiime2 was installed and activated.
Thanks for confirming that. The error is still puzzling to me because it indicates that Dokdo isn't able to find QIIME 2 in the current environment. If you don't mind, could you show me the results of
$ qiime info$ dokdo -vin the command line?
System versions
Python version: 3.8.8
QIIME 2 release: 2021.4
QIIME 2 version: 2021.4.0
q2cli version: 2021.4.0
Installed plugins
alignment: 2021.4.0
composition: 2021.4.0
cutadapt: 2021.4.0
dada2: 2021.4.0
deblur: 2021.4.0
demux: 2021.4.0
diversity: 2021.4.0
diversity-lib: 2021.4.0
emperor: 2021.4.0
feature-classifier: 2021.4.0
feature-table: 2021.4.0
fragment-insertion: 2021.4.0
gneiss: 2021.4.0
longitudinal: 2021.4.0
metadata: 2021.4.0
phylogeny: 2021.4.0
quality-control: 2021.4.0
quality-filter: 2021.4.0
sample-classifier: 2021.4.0
taxa: 2021.4.0
types: 2021.4.0
vsearch: 2021.4.0
for the second command "dokdo -v" gave me the same as previous although the dokdo was installed
after running the command dokdo -v
Traceback (most recent call last):
File "/usr/local/bin/dokdo", line 5, in
from dokdo.main import main
File "/usr/local/lib/python3.8/dist-packages/dokdo/init.py", line 1, in
from .api import *
File "/usr/local/lib/python3.8/dist-packages/dokdo/api/init.py", line 1, in
from .get_mf import get_mf
File "/usr/local/lib/python3.8/dist-packages/dokdo/api/get_mf.py", line 1, in
from qiime2 import Metadata
ModuleNotFoundError: No module named 'qiime2'
I found the problem! You are currently using the 1.0.0 version of Dokdo, which is terribly outdated and doesn't even have the prepare-lefse command. Please re-install the latest version: 1.10.0:
$ git clone https://github.com/sbslee/dokdo
$ cd dokdo
$ pip install .
Please let me know if this doesn't solve the issue.
it is working now
I will go through the remining commands for LEfSe analysis and tell you if the analysis was done perfectly
thank you