Human mutational constraint as a tool to understand biology of rare and emerging bone marrow failure syndromes

Abstract

Inherited bone marrow failure (IBMF) syndromes are rare blood disorders characterized by hematopoietic cell dysfunction and predisposition to hematologic malignancies. Despite advances in the understanding of molecular pathogenesis of these heterogeneous diseases, genetic variant interpretation, genotype–phenotype correlation, and outcome prognostication remain difficult. As new IBMF and other myelodysplastic syndrome (MDS) predisposition genes continue to be discovered (frequently in small kindred studies), there is an increasing need for a systematic framework to evaluate penetrance and prevalence of mutations in genes associated with IBMF phenotypes. To address this need, we analyzed population-based genomic data from >125 000 individuals in the Genome Aggregation Database for loss-of-function (LoF) variants in 100 genes associated with IBMF. LoF variants in genes associated with IBMF/MDS were present in 0.426% of individuals. Heterozygous LoF variants in genes in which haploinsufficiency is associated with IBMF/MDS were identified in 0.422% of the population; homozygous LoF variants associated with autosomal recessive IBMF/MDS diseases were identified in only .004% of the cohort. Using age distribution of LoF variants and 2 measures of mutational constraint, LOEUF (“loss-of-function observed/expected upper bound fraction”) and pLI (“probability of being loss-of-function intolerance”), we evaluated the pathogenicity, tolerance, and age-related penetrance of LoF mutations in specific genes associated with IBMF syndromes. This analysis led to insights into rare IBMF diseases, including syndromes associated with DHX34, MDM4, RAD51, SRP54, and WIPF1. Our results provide an important population-based framework for the interpretation of LoF variant pathogenicity in rare and emerging IBMF syndromes.

Variant interpretation using population databases: Lessons from gnomAD

“…Constraint metrics are key features of the gnomAD database and have been widely used to aid gene and variant interpretation in rare disease (Bamshad et al, 2019; Oved et al, 2020). The constraint metrics are based on a gene's observed versus expected number of very rare SNVs (allele frequency < 0.1%), corrected for sequence context and coverage.…”

Genetic background of primary and familial HLH in Qatar: registry data and population study

“…We can summarize that these genes, classified as highly sensitive to mutations and genetic changes or “mutation and/or genic intolerant” by these measures, should be enriched for variants that lead to genetic diseases compared to genes that are more tolerant to mutations and classified as “ mutation and genic tolerant” ( 62 , 63 ). More precisely, these genes show evidence of intolerance to functional variation (RVIS); however, they appear to be tolerant to putative loss of function variants.…”

Relating pathogenic loss-of function mutations in humans to their evolutionary fitness costs

“…Specifically, the authors assume a Poisson distribution of LOF mutations in a gene and assign an upper 95% confidence limit on the underlying mean number of such mutations, as a factor of the expected number of LOF mutations for this gene (4). Genes classified as highly "mutation intolerant" by these measures are enriched for variants that lead to Mendelian genetic diseases (e.g., (5)(6)(7)(8)(9)). A number of recent papers report an enrichment of variants in "mutation-intolerant genes'' for severe complex disease risk as well (e.g., (10)(11)(12)(13)(14)(15)(16)(17)(18)(19)).…”