Switch error (SE) rates for different methods applied to extended pedigrees.

We evaluate SE for individuals who are members of a complex pedigree (pedigrees that are larger than a parent-child duo and father-mother-child trio). The first row is the number of individuals from each cohort in such pedigrees. The second row shows the yield of SLRP when applied to each cohort. Rows 3–6 show the SE for SHAPEIT2, SLRP, Beagle and HAPI-UR within SLRP detected IBD regions. Rows 7–9 show the SE for SHAPEIT2, Beagle and HAPI-UR outside SLRP detected IBD regions. Rows 10–12 show the overall SE for SHAPEIT2, Beagle and HAPI-UR. Rows 13–15 show the overall SE for SHAPEIT2, Beagle and HAPI-UR haplotypes after correction with the duoHMM method. Row 16 show the overall SE for Beagle applied to pedigrees partitioned into duos and trios where possible. Rows 17–18 show the switch error rate for the SHAPEIT2+duoHMM and Beagle Duo/Trio phasing after masking genotypes flagged as erroneous by the duoHMM.