What is MarkerSet?
MarkerSet is a command-line tool designed for selection of marker panel(s) according to genomic location and known informativity on experimental crosses (e.g. for wide genome scan). MarkerSet can be used with either SNP or microsatellite markers, because it only relies on informativity data. Basically, the algorithm will select the most informative markers in two windows separated by a constant gap, and sliding on the genome. MarkerSet will define automatically the windows size and the gap distance between the two windows depending on the wished number of markers to select and genome size. MarkerSet has several options to optimize markers selection with the possibility to give more weight to the markers distances or to the markers informativity. In case of availability of several experimental designs, it is possible to compare the markers set quality obtained by selecting markers perfectly fitted for each experimental design (monodesign selection), or by selecting a set of markers common to all designs (multidesign selection).
MarkerSet is designed to be compatible with any kind of markers and species.
What do I need to run MarkerSet?
MarkerSet needs a functional PERL environment with POSIX module available (typically *nix system, may work with MacOS X). MarkerSet.pl and config.pm files must be installed in the same directory.
The input file must be a plain tabulated text file with the following informations (headers are requested):
Markers name - chromosome - genomic location - informativity values for cross 1 (typically the number of heterozygote founders) - informativity values for cross 2 …
The available download package comes with core files (PERL program and config file), README file for explanations on program parameters, example input files, and a verbose log file for a better understanding of the algorithm.
Who to contact?
Frédéric Lecerf: lecerf(at)agrocampus-rennes.fr / (33) 2 23 48 59 62
Olivier Demeure: olivier.demeure(at)rennes.inra.fr / (33) 2 23 48 54 66
To download the software use the following link