A web-tool for prioritising genes that regulate the skeleton
This interactive platform is designed to integrate information from (i) a genome-wide association studies (GWAS) of bone mineral density (BMD),
(ii) gene expression data from individual bone and marrow cells isolated from mice,
and (iii) extensive skeletal phenotyping data from human and murine mutant databases.
By triangulating information from these complementary datasets, this platform can help musculoskeletal researchers to systematically prioritise the most likely effector genes
that control BMD and formulate cellular and mechanistic hypotheses regarding their role in skeletal health.
The following example suggests that TNFRSF11A (RANK)
is most likely effector gene at this locus, and that its role in controlling BMD could be mediated by osteoclast function.
The following lines of evidence support this hypothesis: The genetic variant most strongly associated with BMD ( purple diamond ♦ in Panel 1)
intersects the final exon of TNFRSF11A (Panel 4).
Gene based tests show that TNFRSF11A
is more strongly associated with BMD than any other gene in this region (Panel 4) : blue border, p < 2×10-6 | grey border, p < 0.05 - 2×10-6 |
no-border p > 0.05 ). Follow-up in the NGRI GWAS catalogue reveals multiple trait-associations are present at this locus and several intersect
TNFRSF11A (represented by vertical blue lines in Panel 2).
These include pleiotropic associations with Paget’s disease of bone, which is a human disorder characterised by abnormal BMD and altered osteoclast activity. Panel 3 shows that
Tnfrsf11a is differentially expressed in osteoclasts
(represented by a red shaded box that occupies the same genomic co-ordinates as TNFRSF11A ).
None of the other genes at this locus are differentially expressed in any other bone and marrow cells.
Mouse phenotyping studies shows that Tnfrsf11a mouse mutants have abnormal BMD and
osteoclasts (3rd panel – indicated by red colour of TNFRSF11A gene body).
Multiple Phenotypes comparison is underdevelopment