kfarleigh/popgenhelpr

PopGenHelpR is an R package that estimates commonly used population genomic statistics and generates publication quality figures.

popgenhelpr's Issues

Error when knitting rmd file with mapping functions

Users may see the error below when using the Div_Stats_Map or Dif_Stats_Map` functions.

Error in ggplot2::geom_sf():

! Problem while setting up layer.
ℹ Error occurred in the 2nd layer.
Caused by error in sym():
! Can't convert NULL to a symbol.

Error: color grey99 cannot be found

Some users receive an error stating that the color grey99 cannot be found when plotting in PopGenHelpR v1.2.1 or earlier.

Error with mapping functions if rnaturalearthhires not installed.

The functions Plot_ancestry, Dif_Stats_Map, and Div_Stats_Map will throw an error if rnaturalearthhires is not installed. The solution is to install rnaturalearthhires using the code below.

# install devtools if needed
install.packages("devtools")
devtools::install_github("ropensci/rnaturalearthhires")

Error when in ancestry functions when using Q-matrix from LEA

Some users are receiving error messages when using the ancestry functions (Ancestry_barchart and Piechart_map). See the error and solution below.

Errors
Error: object 'Ind_anc' not found
Error: Please supply input for plot.type. The options are 'all', 'individual', or 'population'.

Solution
The functions expect a q-matrix of type data.frame or a csv file where the first column contains sample names, followed by a column for each genetic cluster (K). LEA's Q function does not include sample names in its output and its output is a matrix, not a data.frame or q-matrix. So we need to add sample names and make sure that the columns representing the genetic cluster contribution are numeric.

Code

# If I have a sNMF project named my_project, an optimal K of 3, and my population assignment file is my_pops.csv with the sample names are in the column named Sample.
Qmat <- Q(my_project, K = 3, run = 1)
Loc <- read.csv("my_pops.csv")

# The qmatrix is created correctly; the problem is that it needs to be a data frame for input into the function
# So we add the individual name to the first column (the function needs this to reference the population assignment file)
# But this makes the columns a character which throws another error
# So we make the contribution columns numeric (sapply code)
Qmat_inds <- cbind(Loc$Sample, Qmat)
Qmat_inds_df <- as.data.frame.matrix(Qmat_inds) 

# Make the cluster contributions numeric
Qmat_inds_df[2:4] <- sapply(Qmat_inds_df[2:4], as.numeric)

## Now we can use the Q_mat_inds_df as input
Test_all_ind <- Ancestry_barchart(anc.mat = Qmat_inds_df, pops = Loc, K = 3 ,plot.type = 'individual', col = c('red','blue','purple'))
          
Test_all_pop <- Ancestry_barchart(anc.mat = Qmat_inds_df, pops = Loc, K = 3 ,plot.type = 'all', col = c('red','blue','purple'))

Please reach out if you have questions!