Problem with plot_cladogram

[Claude Stevenon]

Hi,

I have run a LEFSE analysis and found a problem with the output of plot_cladogram. In this plot there should be the same taxonomic levels as in the file used to do the analysis. In my case I have six levels Domain/Phylum/Class/Order/Family/Genus but plot_cladogram shows 7 leves (see attached file). Maybe there is a problem with the format of my data? Another question is that I found the same results using the strict (-y 1) and "no strict" version with run_lefse. Is this possible?

Thank you very much for your help


class    cast    cast    cast    cast    cast    cast    cast    cast    cast    cast    cast    cast    manure    manure    manure    manure    manure    manure    manure    manure    manure
subclass    cab    cab    cab    cab    cab    vac    vac    vac    vac    cer    cer    cer    cab    cab    cab    vac    vac    vac    cer    cer    cer
id    worm01    worm04    worm06    worm07    worm08    worm164    worm165    worm167    worm168    worm84    worm87    worm90    worm_A13    worm_A14    worm_A16    worm_A18    worm_A19    worm_A20    worm_A23    worm_A24    worm_A25
Bacteria|Proteobacteria|Gammaproteobacteria|Pseudomonadales|Moraxellaceae|Acinetobacter|    376    201    305    420    316    0    33    0    158    93    307    159    144    54    50    6    2    13    0    0    0

python format_input.py lefse_rarefotus.txt lefse.in -c 1 -s 2 -u 3 -o 1000000
python run_lefse.py lefse.in lefse.res
python plot_res.py lefse.res lefse.svg --format svg --dpi 300
python plot_cladogram.py lefse.res lefse.cladogram.svg --format svg --dpi 300

[Nicola Segata]

Hi Claude,
 with "--max_lev" you can change the number of taxonomic levels to display, so --max_lev 6 should do the trick. The -y 1 option acts on how classes are compared for datasets with more than two classes which does not apply to your dataset and it thus has no effect.

hope this helps

cheers

Nicola

[Claude Stevenon]

Hi Nicola,

Thanks very much for your help. If I understand your response I can use more than one class, in a "similar" way to factors in a conventional lineal model?

Cheers