LAUSR

MOTIVATION Linkage mapping provides a practical way to anchor de novo genome assemblies into chromosomes and to detect chimeric or otherwise erroneous contigs. Such anchoring improves with higher numbers of markers and individuals, as long as the mapping software can handle all the information. Recent software Lep-MAP3 can robustly construct linkage maps for millions of genotyped markers and on thousands of individuals, providing optimal maps for genome anchoring. For such large data sets, automated and robust genome anchoring tool is especially valuable and can significantly reduce intensive computational and manual work involved. RESULTS Here we present software Lep-Anchor to anchor genome assemblies automatically using dense linkage maps. As the main novelty, it takes into account the uncertainty of the linkage map positions caused by low recombination regions, cross type or poor mapping data quality. Furthermore, it can automatically detect and cut chimeric contigs, and use contig-contig, single read or alternative genome assembly alignments as additional information on contig order and orientations and to collapse haplotype contigs.We demonstrate the performance of Lep-Anchor using real data and show that it outperforms ALLMAPS on anchoring completeness and speed. Accuracy-wise Lep-Anchor and ALLMAPS are about equal, but at the expense of lower completeness of ALLMAPS. The software Chromonomer was faster than the two other methods but has major limitations and is lower in accuracy. We also show that with additional information, such as contig-contig and read alignments, the anchoring completeness can be improved by up to 70% without significant loss in accuracy. Based on simulated data, we conclude that the anchoring accuracy can be improved by utilising information about map position uncertainty. Accuracy is the rate of contigs in correct orientation and completeness is the number contigs with inferred orientation. AVAILABILITY Lep-Anchor is available with the source code under GNU general public license from http://sourceforge.net/projects/lep-anchor. All the scripts and code used to produce the reported results are included with Lep-Anchor.