A systematic genetic analysis and visualization of phenotypic heterogeneity among orofacial cleft GWAS signals

Abstract

Phenotypic heterogeneity is a hallmark of complex traits, and genetic studies of such traits may focus on them as a single diagnostic entity or by analyzing specific components. For example, in orofacial clefting (OFC), three subtypes -cleft lip (CL), cleft lip and palate (CLP), and cleft palate (CP) have been studied separately and in combination. To further dissect the genetic architecture of OFCs and how a given associated locus may be contributing to distinct subtypes of a trait we developed a framework for quantifying and interpreting evidence of subtype-specific or shared genetic effects in complex traits. We applied this technique to create a "cleft map" of the association of 30 genetic loci with three OFC subtypes. In addition to new associations, we found loci with subtype-specific effects (e.g., GRHL3 (CP), WNT5A (CLP)), as well as loci associated with two or all three subtypes. We crossreferenced these results with mouse craniofacial gene expression datasets, which identified additional promising candidate genes. However, we found no strong correlation between OFC subtypes and expression patterns. In aggregate, the cleft map revealed that neither subtype-specific nor shared genetic effects operate in isolation in OFC architecture. Our approach can be easily applied to any complex trait with distinct phenotypic subgroups.

Pleiotropy method identifies genetic overlap between orofacial clefts at multiple loci from GWAS of multi-ethnictrios

“…Additionally, the 1p36.13 (PAX7), 3q12.1 (COL8A1), 8q21.3 (DCAF4L2) and 8q24 (gene desert) regions with shared effects between CL and CLP appear to have at least 2 distinct loci of genetic overlap indicated by more than one peaks ( Figure S16). Presence of possibly independent loci in the 8q24 region has been reported previously 14 . Furthermore, PLACO revealed a novel OFC locus at 1p21.3 (MIR137HG, p = 2.2 × 10 −8 ) with opposite genetic effects for CL & CLP at the genome-wide threshold ( Table 1 and Figure S17 a).…”

“…The copyright holder for this this version posted November 16, 2020. ; https://doi.org/10.1101/2020.11.13.20231654 doi: medRxiv preprint both shared and opposite effects at IRF6 has been reported previously 14,30 . panels f, g).…”

“…The 2 loci in LIMCH1 and RAB8A are novel for OFCs. All the other genes have been implicated in GWAS of CL/P previously 2,14 , and insights into the molecular pathogenesis of OFCs via many 4 . CC-BY-ND 4.0 International license It is made available under a is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity.…”

“…When investigating the RR estimates from the top several SNPs at each of the above-mentioned loci shared by CL and CLP, we found loci suggesting different pathogenesis of these subtypes at 1p22.1 (ABCA4, ARHGAP29) and 1q32.2 (IRF6) as evidenced by a large number of SNPs with opposite genetic effects ( Figure S15). Some previous studies 14,33 have noted differences in genetic etiologies of CL and CLP, particularly at the IRF6 locus. Additionally, the 1p36.13 (PAX7), 3q12.1 (COL8A1), 8q21.3 (DCAF4L2) and 8q24 (gene desert) regions with shared effects between CL and CLP appear to have at least 2 distinct loci of genetic overlap indicated by more than one peaks ( Figure S16).…”

“…Furthermore, if a locus is hypothesized to have opposite genetic effects on the two subtypes (e.g., NOG 14,19 ), the pooling technique will dilute any signal and consequently will be under-powered to detect genetic overlap. Use of multi-trait methods in OFC genetic studies, such as the ones commonly used in population-based GWAS of complex traits [23][24][25][26] , is hindered by the disjoint nature of the subtypes (i.e., absence of subjects with both traits), the qualitative (binary) nature of the traits, and/or the case-parent trio design typically used to study multi-ethnic samples with OFC.…”