Read GeneMapper Genotypes Tables
Given a vector of filepaths to tab-delimited text files containing
genotype data in the ABI GeneMapper Genotypes Table format,
read.GeneMapper produces a
genambig object containing
the genotype data.
A character vector of paths to the files to be read.
read.GeneMapper can read the genotypes tables that are exported
by the Applied Biosystems GeneMapper software. The only alterations
to the files that the user may have to make are 1) delete
any rows with missing data or fill in
-9 in the first allele slot for
that row, 2) make sure that all allele names are numeric
representations of fragment length (no question marks or dashes), and
3) put sample names into the Sample Name column, if the names that you
wish to use in analysis are not already there. Each file should have
the standard header row produced by the software. If any sample has
more than one genotype listed for a given locus, only the last
genotype listed will be used.
The file format is simple enough that the user can easily create files
manually if GeneMapper is not the software used in allele calling.
The files are tab-delimited text files. There should be a header row
with column names. The column labeled “Sample Name” should contain
the names of the samples, and the column labeled “Marker” should
contain the names of the loci. You can have as many or as few columns as
needed to contain the alleles, and each of these columns should be
labeled “Allele X” where X is a number unique to each column. Row
labels and any other columns are ignored. For any given sample, each
allele is listed only once and is given as an integer that is the
length of the fragment in nucleotides. Alleles are separated by
tabs. If you have more allele columns than alleles for any given
sample, leave the extra cells blank so that
read them as
NA. Example data files in this format are
included in the package.
read.GeneMapper will read all of your data at once. It takes
as its first argument a character vector containing paths to all of
the files to be read. How the data are distributed over these files
does not matter. The function finds all unique sample names and all
unique markers across all the files, and automatically puts a missing
data symbol into the list if a particular sample and locus combination
is not found. Rows in which all allele cells are blank should NOT be
included in the input files; either delete these rows or put the
missing data symbol into the first allele cell.
Sample and locus names must be consistent within and across the files. The object that is produced is indexed by these names.
forceInteger=FALSE, alleles can be non-numeric values. Some
functionality of polysat will be lost in this case, but it could
allow for the import of SNP data, for example.
genambig object containing genotypes from the files, stored as
vectors of unique alleles in its
Genotypes slot. Other slots are
left at the default values.
A ‘subscript out of bounds’ error may mean that a sample name
or marker was left blank in one of the input files. A ‘NAs
introduced by coercion’ warning when
forceInteger=TRUE means that a
non-numeric, non-whitespace character was included in one of the
allele fields of the file(s), in which case the file(s) should be
carefully checked and re-imported.
Lindsay V. Clark
GeneMapper website: https://www.thermofisher.com/order/catalog/product/4475073
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18
# create a table of data gentable <- data.frame(Sample.Name=rep(c("ind1","ind2","ind3"),2), Marker=rep(c("loc1","loc2"), each=3), Allele.1=c(202,200,204,133,133,130), Allele.2=c(206,202,208,136,142,136), Allele.3=c(NA,208,212,145,148,NA), Allele.4=c(NA,216,NA,151,157,NA) ) # create a file (inspect this file in a text editor or spreadsheet # software to see the required format) write.table(gentable, file="readGMtest.txt", quote=FALSE, sep="\t", na="", row.names=FALSE, col.names=TRUE) # read the file mygenotypes <- read.GeneMapper("readGMtest.txt") # inspect the results viewGenotypes(mygenotypes)
Want to suggest features or report bugs for rdrr.io? Use the GitHub issue tracker.