Mapping non-host resistance to the stem rust pathogen in an interspecific barberry hybrid

BMC Plant Biology

Table 1 Description of the sequence of filters applied to obtain the final marker sets for linkage map construction

Filter descriptions, in order of application	Markers removed at each step	Markers retained
1. More than 30% missing genotype calls across the population^a	6106	9305
2. Heterozygous in both parents	272	9033
3. Homozygous for alternate alleles in the two parents^b	3982	5051
4. Deviates significantly from expected allele depth ratio in heterozygotes^c	1801	3250
5. Segregating genotypes unsupported by parental genotypes^d	697	2553
6. Deviates significantly from expected Mendelian segregation^e	90	2463
Final markers for the B. thunbergii linkage map (Bt × Bv)		1757
Marker Set 1: ab × aa		1497
Marker Set 2: cd × −-		260
Final markers for the B. vulgaris linkage map (Bt × Bv)		706
Marker Set 3: ee × ef		600
Marker Set 4: -- × gh		106

^a This first filter was applied to the initial set of 15,411 markers (SNPs and indels) identified by the GBS-SNP-CROP pipeline
^b If both parents are homozygous for the marker, no variation will be observable among the progeny (i.e. all F₁ progeny will be heterozygous for the marker)
^c Mean allele depth ratio across heterozygous F₁ progeny deviates > 25% from the expected bi-allelic depth ratio of 1:1
^d Lack of parental genotypes (missing data) and/or parental genotyping errors can prevent the unique assignment of gametic origin. For example, while ab × aa is expected to segregate only as aa and ab among the progeny, the alternate homozygote (bb) may be observed due to parental genotyping error. All such markers were removed from the analysis
^e Segregation ratio of genotypes deviates more than two standard deviations from the expectation for each marker set; such markers were removed due to their high segregation distortion

ISSN: 1471-2229