Fig. 4

Total urinary volatile phenotypes do not correspond with MUP genotype. A ESI-MS intact mass profiles compare central MUP profiles among trios of males in four sib groups (a−d, background shading), where sib groups were double cousins (different parents, shared grandparents; r = 0.25). Within sib trios (r = 0.5), two brothers shared both MUP haplotypes, whilst a third shared only one haplotype with sibs (haplotypes denoted by capital letters). B Pearson correlation coefficients compare similarity of MUP profiles between each pair of males according to shared MUP haplotypes and level of relatedness, based on the relative proportion of each of seven mass peaks in their intact mass profiles (data are means ± SEM). Each mass peak corresponds to known mouse MUP sequences (see Data analysis section in Methods). C Pearson correlation coefficients also compare the similarity of total urinary volatile profiles between pairs of males, based on the areas of 134 peaks extracted by HS-SPME and GC-MS analysis, expressed as proportions of the total peak area in that sample (data are means ± SEM). D Principal component analysis of urinary volatile phenotypes reveals clustering of males within sib groups (symbols) based on the first three derived components (explaining 24.6%, 12.7% and 9.5% variance respectively). Two urine samples were analysed from male 1 of each sib group (symbols linked by lines) to compare similarity within and between males. There was no clear clustering according to MUP genotype (colours)