search

nhanhocu / metamicrobiomeR

An R package for analysis of microbiome relative abundance data using zero inflated beta GAMLSS and meta-analysis across microbiome studies using random effects models
https://github.com/nhanhocu/metamicrobiomeR
33 stars 11 forks source link

problems with taxa.mean.plot and different tax. levels #3

Open VGalata opened 5 years ago

VGalata commented 5 years ago

Hi,

I encountered some problems when using taxa.mean.plot:

  1. tax.level="l2": Here, entries without phylum are not removed, .g. I still get k__archaea and k_bacteria in my plot. Also, this seems to compute wrong values for the bar plot as the lower and upper parts are together more than 100% in my case.
  2. tax.level="l3": Here, an error is returned: Error in paste(".*", stringt, sep = "") : object 'stringt' not found. Since, there is no if clause for this value stringt is not defined before it is used in line 93.
nhanhocu commented 4 years ago

Hi, Sorry for not being up-to-date with the comments, issues. Regarding the issue, you should use taxa.filter to select the list of taxa you would like to plot (usually you can use abundance and prevalance threshold to remove unwanted one like k__archaea ). For the list of taxa that you would like to plot after filtering, you can put that as argument for tax.select in the taxa.mean.plot function. You are right that stringt was not defined for "l3". This is a bug, sorry about that. I am revising the "metamicrobiomeR" package to put in R cran and will fix these issues and update the package all together.

nhanhocu commented 4 years ago

Hi, Please try our new version in R cran in which the above issue have been fixed. Thanks, Nhan