CrossLink: Genetic Mapping Software for Outbreeding Species

Genetic mapping plays a central role in the study and breeding of plants, facilitating scaffolding of full chromosome-length genome sequences, interspecies synteny comparisons and quantitative trait loci identification.

Map construction is relatively straightforward where a traditional backcross or F2 population from inbred lines can be used. Some species, including many crops, suffer inbreeding depression from the recessive deleterious alleles present in outbreeding populations. To avoid excessive segregation distortion, maps must therefore be produced from outcrosses between highly heterozygous parents, making more specialized mapping approaches necessary.

We present CrossLink, software for the rapid and automated construction of genetic maps for outbreeding species, using biallelic markers such as can be obtained from SNP arrays, genotyping-by-sequencing or whole genome sequencing.

Incorporating the best methods available from existing mapping programs, our software groups, phases and orders markers using algorithms designed to scale efficiently to dense maps. Initial ordering uses the fast minimum spanning tree method of the (outcross-incompatible) MSTmap program. Final map ordering and multipoint map distance estimation are implemented using a genetic algorithm and Gibbs sampler respectively, similar to the methods employed by JoinMap(R). The whole mapping process is designed to be suitable for automation and scripting in a UNIX environment. The software was tested on simulated data and SNP array data from an outcross between Redgauntlet and Hapil octoploid strawberry cultivars.

Our source code will be made available on github under an open source licence to facilitate customization and improvement by the research community.