developed the pLI score that reflects each gene's intolerance to loss-of-function variation. 2016) currently provide data on more than 100,000 exomes from diverse populations combined. Databases such as ExAC or gnomAD ( Lek et al. Hence, one of the most active areas of development has been the ongoing expansion and development of large exome resources. 2016) or family-based sequencing data such as trio sequencing, particularly in the context of de novo variant discovery ( Zhu et al. However, in both scenarios, identifying which variants are most likely to be pathogenic has required either large control data sets ( Dewey et al. Subsequent filtering of these candidate variants to identify disease genes is generally easier for recessive inherited diseases than dominant inherited diseases in which one causal variant is expected instead of two. 2016) among which four are splice disrupting. The average individual carries approximately 20 rare, “loss-of-function” variants (MAF ≤ 0.1) ( Dewey et al. Without prior biological knowledge, exome sequencing a single case has a low probability of yielding a novel pathogenic variant or gene. Without knowing the cause of the disease, it can be challenging to select a complementary approach in the postexome “experimental maze.” Depending on the causal variant type, different options are available, influenced by both technical and biological diagnostic challenges. 2018).Ĭhallenges and approaches in “exome-negative” cases. 2018) and in cases in which it is not exonic ( Short et al. 2018) if it is a structural variant not properly caught through exome sequencing ( Merker et al. 2014 Hoffman-Andrews 2017) if multiple variants are responsible for the pathology, in two or more genes ( Liu et al. 2016): If it is in coding regions but not properly detected ( Cornish and Guda 2015) if it is of unknown significance and was not selected ( McCarthy et al. To summarize, the causal variant can be missed in cases of somatic mosaicism ( Priest et al.
Negative exome results can be explained in various ways ( Fig. For “exome-negative” cases, in which no such diagnosis is provided, there remain multiple approaches after the exome-but choosing that next step in an “experimental maze” still remains a major challenge.
2018) can be more challenging to perform than well described-although rare-pathologies ( Aartsma-Rus et al. However, diagnoses of diseases involving novel mechanisms ( Oláhová et al. In clinical settings, depending on disease type and patient selection, exome sequencing has been estimated to lead to a diagnosis in 30%–50% of rare Mendelian diseases ( McInerney-Leo et al. 2016 Deciphering Developmental Disorders Study 2017). For patients with shared rare diseases, it has aided in identifying novel causal genes by their excess of pathogenic variants ( Ng et al. High-throughput sequencing has helped to uncover the rates at which deleterious or damaging alleles accumulate in specific genes ( Lek et al.