Construction of a 10,000-Marker Ultradense Genetic Recombination Map of Potato: Providing a Framework for Accelerated Gene Isolation and a Genomewide Physical Map Hans van Os,* Sandra Andrzejewski, Erin Bakker, Imanol Barrena, ､ Glenn J. Br yan,** Bernard Caromel, Bilal Ghareeb, Edwige Isidore,** Walter de Jong,**,1 Paul van Koert, 2 Ve ́ ronique Lefebvre, Dan Milbourne,**, 3 Enrique Ritter, Jeroen N. A. M. Rouppe van der Voort, 2 Francxoise Rousselle-Bourgeois, Joke van Vliet, Robbie Waugh,** Richard G. F. Visser,* Jaap Bakker and Herman J. van Eck*, 4

Main Goal What? –Construct an ultradense genetic linkage map in potato Results in global saturation of the genome with marker loci Why? –Useful for mapping applications –Marker distances less than BAC inserts so chromosome landing is easier –Facilitates genetic anchoring of a physical map –Useful for research community and can be transferable to other potato genotypes or populations

Associated problems with creating ultradense linkage maps Computer programs for linkage mapping are unable to handle such vast amounts of data Scoring errors result in high rates of ordering ambiguities between markers within short genetic distances

Materials and Methods Plant Material –2 diploid heterozygous potato clones crossed to make 136 F 1 individuals Marker Analysis –AFLP markers were made using EcoRI/MseI, SacI/MseI, PstI/MseI combinations with 2 and 3 selective nucleotides respectfully (381 primer combinations) –Linkage groups assigned to 12 potato chromosomes using a set of AFLPs with known positions

Materials and Methods Map construction –Markers split into 3 types Heterozygous for maternal, paternal or both –Marker order and linkage phase determined by JMQAD32 (JoinMap2.0) Quick and dirty –Order of markers reordered with RECORD Aligns markers in most parsimonious order using the cost function Displayed as graphical genotype in Excel so ambiguous data points could be discarded –Corrected data ordered again with RECORD and singletons removed with SMOOTH –ComBin removes redundancy caused by cosegregating markers Once ComBin displays a linear figure it is concluded a linkage group is free from data ambiguities Bins are then filled with marker loci

Figure 1. Overview of Method

Results 10,365 scorable markers recorded – markers/linkage group 12 maternal and paternal skeleton maps deduced from bin signatures –1 bin = 100/130cM Scoring data was fitted into the bins of the skeleton bin map using maximum likelihood

Strong congruence between maternal and paternal maps Dark regions indicate numerous markers and signify proposed centromere locations Appear to be crossover hotspots and cold spots

Conclusions Greater than 10,000 markers generated at a proposed 93% accuracy rate Strong correlation between maternal and paternal maps with respect to map length and position of strong clustering Marker gaps likely to be due to recombination hotspots or local fixation Strong chiasma interference evident after comparing observed and expected chiasma Underrepresentation of PstI markers at proposed centromeres because PstI cuts at hypomethalyated gene-rich regions Markers found at