A Statistical Method to Establish the Presence of Oligogenicity in Isolated GnRH Deficiency: Use of Exac Population Database and Insights of Relevance to Human Mendelian Disorders

Presentation Number: SUN 468
Date of Presentation: April 2nd, 2017

Riwa Sabbagh*1, Ravikumar Balasubramanian1, Monkol Lek1, Edward R. Horton1, Nada Al Tassan2, Dorota Monies2, Lacey Plummer1, Stephanie Beth Seminara3, Brian Meyers2, Mark Daly1, Daniel MacArthur1 and William F Crowley4
1Massachusetts General Hospital, Boston, MA, 2Department of Genetics at the Research Centre of King Faisal Specialist Hospital and Research Centre, Riyadh, Saudi Arabia, 3Mass Gen Hosp, Boston, MA, 4Mass. General Hospital


Introduction: Isolated GnRH Deficiency [IGD] is a rare genetic disorder causing pubertal failure. In addition to Mendelian inheritance, oligogenicity has been reported in IGD (Sykiotis, PNAS, 2010). To-date, only a subset of IGD genes have been systematically examined for oligogenicity and a robust statistical validation of oligogenicity in human genetic disorders is lacking. Utilizing the power of the Exome Aggregation Consortium (ExAC) population dataset, a comparative genetic burden analysis for oligogenicity involving an expanded list of 13 causal autosomal IGD genes was performed.

Methods:  703 Non-Finnish Europeans (NFE) patients underwent Sanger sequencing for 13 known IGD genes (FGF8/FGFR1, GNRH1/GNRHR, HS6ST1, KISS1/KISS1R, NSMF, PROK2/PROKR2, and TAC3/TACR3). Sequence data for these genes was extracted from ExAC in 33,370 NFE controls. Rare (minor allele frequency [MAF] <1%) synonymous variants (SVs) and protein-altering non-synonymous variants (NSVs) [loss-of-function [LOF] (stop, frameshift, splice); missense] were compiled for both groups. An expected and observed burden for oligogenic hits (NSVs in >2 genes) was first performed within ExAC to validate it as an appropriate control cohort. A step-wise burden analysis, IGD vs. ExAC, was then performed across three MAF bins (<1%, <0.5%, <0.1%) for: (i) SV mutation burden; (ii) NSV mutation burden; and finally, (iii) oligogenicity burden [Fischer’s exact test].

Results:  The observed oligogenic counts in ExAC were not different from the expected statistical probability and the SV burden between IGD and ExAC was identical, validating the use of ExAC as an appropriate control group. However, the IGD cohort was significantly enriched for a cumulative NSV burden (p<0.001) as well as a cumulative oligogenicity burden across all MAF bins (MAF <1%: p=0.005, MAF <0.5%: p=0.006, MAF<0.1%: p=0.002). Intriguingly, when considering the nature of oligogenic NSVs, the IGD cohort was only significantly enriched when the oligogenic pair contained at least one LOF variant (MAF <1%: p=3.23x10-7, MAF <0.5%: p=6.28x10-8; MAF <0.1%: p=1.18x10-5. In contrast, oligogenic pairings comprised of exclusively missense variants showed a trend towards significance only for ultra-rare variants [MAF<0.1%; p=0.07].

Conclusions: 1) A new analytical framework to validate oligogenicity in IGD using a large control cohort is now available and can now be applied to determine the presence of oligogenicity in other human mendelian disorders; 2) Oliogogenic pairs with at least one LOF variant were significantly enriched in IGD while exclusively missense pairs only approached statistical significance when variants were ultra-rare; and 3) These observations highlight the importance of considering the precise nature and allele frequency of oligogenic pairings to ascertain causality and guide appropriate genetic counseling for IGD patients.


Nothing to Disclose: RS, RB, ML, ERH, NA, DM, LP, SBS, BM, MD, DM, WFC