read.GeneMapper: Read GeneMapper Genotypes Tables

Description Usage Arguments Details Value Note Author(s) References See Also Examples

View source: R/dataimport.R


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.


read.GeneMapper(infiles, forceInteger=TRUE)



A character vector of paths to the files to be read.


Boolean. If TRUE, alleles will be coerced to integers. This is particularly useful for stripping any white space from allels and preventing alleles from being imported as character strings. If FALSE, alleles will be imported as numeric or character values, depending on the content of the input file(s).


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. Duplicate alleles in the same row are ignored by read.GeneMapper. 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.table will 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.

If 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.


A 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:

See Also

genambig, read.Structure, read.GenoDive, read.SPAGeDi, read.Tetrasat, read.ATetra, write.GeneMapper, read.POPDIST, read.STRand


# create a table of data
gentable <- data.frame(Sample.Name=rep(c("ind1","ind2","ind3"),2),
                       Marker=rep(c("loc1","loc2"), each=3),
# 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

polysat documentation built on June 8, 2018, 1:04 a.m.