Expanded functionality, increased accuracy, and enhanced speed in the de novo genotyping-by-sequencing pipeline GBS-SNP-CROP

Note: In total, 25 000 SNPs and 10 000 indels were simulated across a genomic space of 100 000 GBS fragments. A total of 60 002 165 single-end reads were simulated for a population of 25 individuals (average of 2.4 million reads per genotype), with a sequencing error rate of 1.1%. See Supplementary Table S1 for more details UNEAK 1⁄4 TASSEL-UNEAK; GSC 1⁄4 GBS-SNP-CROP. The number of genotypes used for mock reference (MR) assembly. Computation time (minutes) required to run the full analysis on a Unix workstation with 16 GB RAM and a 2.6 GHz Dual Intel processor. Number of variants called by a pipeline (Note: a total of 35 000 variants were simulated, consisting of 25 000 SNPs and 10 000 indels). Percentage of called variants that could not be validated (false positives). Percentage of true, simulated variants that were not detected by the pipeline. Overall accuracy: 100 * [number of validated variants/(total number of simulated variants þ number of non-validated variants)].