Skip to main content
Fig. 5 | BMC Bioinformatics

Fig. 5

From: A joint use of pooling and imputation for genotyping SNPs

Fig. 5

Genotypes imputation accuracy in a classical and a pooled scenario. a and b concordance (based on best-guess genotype) c and d cross-entropy (based on posterior genotypes probability) metrics. All markers from the HD map have been used for computing the metrics (52,697 markers). Beagle (labeled as “beagle”) performance is in blue, and Prophaser (labeled as “phaser”) in orange. The central line is the median and the shadowed areas delimit the percentiles 0.0, 0.01, 0.25, 0.75, 0.99, 1.0. The x-axis was built from 0.05-long MAF bins within which each marker concordance score was computed as the mean score of the 500 previous and 500 next markers sorted per ascending MAF

Back to article page