synthetic.cross.RM data frame exported by match.G2A in which combinations of row and
columns were being replicated.Added a NEWS.md file to track changes to the package.
Small internal changes for better integration with ASRgwas.
Enabled qc.filtering to finish calculations when a single marker is left.
Code synthetic.cross will leave only one of the records if duplicated in the pedigree.
Code kinship.diagnostics will leave only one of the records if duplicated in the kinship matrix.
kinship.diagnostics with clean = TRUE will not crash if there is nothing to remove.The function snp.pruning for marker pruning based has been added.
The function synthetic.cross for obtaining the genotypes of offspring off of a set of crosses has been added.
The function snp.recode for marker recoding has been added.
The function H.matrix, which is a simplification of H.inverse, has been added.
In kinship.diagnostics, an option to determine the proportion of data points to use for plotting has been added to improve processing time.
Two new filters for markers have been added to qc.filtering, one based on inbreeding and one based on the observed heterozygosity (suggested by Simon Nadeau).
An option to draw group-dependent ellipses has been added to both snp.pca and kinship.pca.
Now the markers map can be passed to qc.filtering and will also be filtered.
Relevant attributes of an object (such as INVERSE) are preserved after using full2sparse and sparse2full.
Now qc.filtering can better deal with markers with all samples missing. These markers are removed first thing when the function is called.
Plot generation in match.G2A processing time has been reduced by a factor of ~3.
The recoding section of qc.filtering has been put into another function (snp.recode). The arguments related to recoding in the first have not been removed, but if used will generate a stop suggesting the usage of the latter. Consequently, the M.recode output has been removed.
More message control with message = TRUE/FALSE has been added to several functions.
Some internal linear algebra has been changed to improve processing time.
Error handling has been improved in G.inverse to provide more informative messages to the user.
Help files have been reviewed and improved.
colNames attribute is correctly assigned in full2sparse.