Mobile Element Insertion Detection in Genome Sequencing of a Cohort of Unsolved Cases
Laboratory Genetics and Genomics
-
Primary Categories:
- Laboratory Genetics
-
Secondary Categories:
- Laboratory Genetics
Introduction:
Mobile element insertions (MEIs) are an increasingly recognized driver of genomic alteration which may result in disease pathogenesis in cancer and germline genetic disorders. MEIs are repetitive sequences ~100-6000 bp in length that constitute a substantial portion of the human genome. Most have lost the ability to move within the genome, though an important fraction of MEIs have retained that ability and are able to insert into novel positions that may impact disease. Because MEIs are repetitive and ubiquitous throughout the human genome, they are frequently missed in standard short-read NGS pipelines due to incorrect read mapping. Previous studies of MEI detection in exome sequences have reported diagnostic rates of 1 in 3000-4000. In this study, we utilized a specialized tool to detect MEIs called the Mobile Element Locator Tool (MELT) on a cohort of 149 unsolved rare disease cases and demonstrate that mobile element insertions may be an important cause of rare diseases due to disruption of exonic and intronic regions.
Methods:
We performed genome sequencing (GS) for 149 unsolved rare disease cases that had previously been undiagnosed by exome sequencing. Of those, we also performed RNA sequencing on 148 of those cases. MELT was chosen because it was developed for GS and used by gnomAD. We analyzed MEI calls that were present in known disease-associated genes (OMIM as of Nov 2024), or genes of uncertain significance (GUS) with gnomAD pLI > 0.9. Additionally, we filtered for calls that were unique within our cohort, absent from homozygotes in gnomAD, and present in <10 heterozygotes for autosomal dominant genes or GUSs.
Results:
We identified 370 MEI calls (2.5 per proband) that met our filtering criteria. Manual review of these insertions revealed 6 that were present in genes highly relevant to the proband’s phenotype. Of these events, 1 was exonic, 2 were in intronic regions within 50 bp of the splice junction, and 3 were deep intronic. Analysis of the RNA-seq data for the corresponding probands showed that 4 of these events were in genes with no or poor coverage in blood. For the 2 events with coverage, 1 did not show aberrant events in the RNA-seq and the other showed a significant decrease in YEATS2 expression by outlier analysis (z-score = -1.7, p-val = 0.02) in a proband with cortical myoclonus.
Conclusion:
We have identified MEIs affecting highly relevant genes in 6 probands, representing a potential diagnostic rate of 1:62, indicating that MEI detection in genome sequencing may lead to more diagnoses than in exome sequencing. However, it will be critical to further characterize the impact of these MEIs to determine if they are pathogenic. Toward that end, we plan to first confirm the MEIs with Sanger sequencing, then query the impact of the MEIs with mini-gene assays for the intronic insertions or assays that measure impact to the specific gene for the exonic insertions.
Mobile element insertions (MEIs) are an increasingly recognized driver of genomic alteration which may result in disease pathogenesis in cancer and germline genetic disorders. MEIs are repetitive sequences ~100-6000 bp in length that constitute a substantial portion of the human genome. Most have lost the ability to move within the genome, though an important fraction of MEIs have retained that ability and are able to insert into novel positions that may impact disease. Because MEIs are repetitive and ubiquitous throughout the human genome, they are frequently missed in standard short-read NGS pipelines due to incorrect read mapping. Previous studies of MEI detection in exome sequences have reported diagnostic rates of 1 in 3000-4000. In this study, we utilized a specialized tool to detect MEIs called the Mobile Element Locator Tool (MELT) on a cohort of 149 unsolved rare disease cases and demonstrate that mobile element insertions may be an important cause of rare diseases due to disruption of exonic and intronic regions.
Methods:
We performed genome sequencing (GS) for 149 unsolved rare disease cases that had previously been undiagnosed by exome sequencing. Of those, we also performed RNA sequencing on 148 of those cases. MELT was chosen because it was developed for GS and used by gnomAD. We analyzed MEI calls that were present in known disease-associated genes (OMIM as of Nov 2024), or genes of uncertain significance (GUS) with gnomAD pLI > 0.9. Additionally, we filtered for calls that were unique within our cohort, absent from homozygotes in gnomAD, and present in <10 heterozygotes for autosomal dominant genes or GUSs.
Results:
We identified 370 MEI calls (2.5 per proband) that met our filtering criteria. Manual review of these insertions revealed 6 that were present in genes highly relevant to the proband’s phenotype. Of these events, 1 was exonic, 2 were in intronic regions within 50 bp of the splice junction, and 3 were deep intronic. Analysis of the RNA-seq data for the corresponding probands showed that 4 of these events were in genes with no or poor coverage in blood. For the 2 events with coverage, 1 did not show aberrant events in the RNA-seq and the other showed a significant decrease in YEATS2 expression by outlier analysis (z-score = -1.7, p-val = 0.02) in a proband with cortical myoclonus.
Conclusion:
We have identified MEIs affecting highly relevant genes in 6 probands, representing a potential diagnostic rate of 1:62, indicating that MEI detection in genome sequencing may lead to more diagnoses than in exome sequencing. However, it will be critical to further characterize the impact of these MEIs to determine if they are pathogenic. Toward that end, we plan to first confirm the MEIs with Sanger sequencing, then query the impact of the MEIs with mini-gene assays for the intronic insertions or assays that measure impact to the specific gene for the exonic insertions.