CADD Scores Research Articles

A family‐based genetic study identifies mutations in TLR9 impairing receptor activation: A role for innate immunity in AD pathogenesis

AbstractBackgroundFamily‐based genetic studies in early‐onset Alzheimer disease (EOAD), prompting early gene discoveries, remain powerful for the identification of novel genes in AD. Causal mutations in APP, PSEN1 and PSEN2 explained <10% of EOAD patients, while several EOAD families remained genetically unexplained. We identified a multigenerational Flanders‐Belgian family with autosomal dominant EOAD. DNA, biomaterials and clinical data from two consecutive informative generations were available: 5 AD patients (onset 64.2±13.2 years, range 52‐85) of which 2 with a definite neuropathological diagnosis and 9 unaffected/at‐risk individuals.MethodWhole exome sequencing of 4 patients. Selection of shared rare non‐synonymous variants, validation of co‐segregation with disease and variants frequency assessment using massive parallel‐targeted sequencings (MPS). We used genotyping of STR markers to define haplotypes. Coding regions of candidate genes were sequenced using MPS in a Belgian cohort of 320 EOAD (onset 57.8±5.9y, range 33‐65) and 1166 late‐onset (LO)AD (onset 77.1±6.5y, range 66‐96) patients and a control cohort of 934 persons (inclusion 70.45±9.2y, range 52‐100). Variants pathogenicity was predicted using CADD scores cutoff>20. Evaluation of functional effects by NF‐κB luciferase assay.ResultThree rare variants co‐segregated with AD in the family: CCR3 p.F249Hfs*23 (rs561062190), ZNF589 p.T355M (rs376706270) and TLR9 p.E317D (novel) and are located within a co‐segregating haplotype block of 15.3 cM at ch3p21. Only the TLR9 variant was absent in controls. TLR9 is a DNA‐sensing receptor expressed in immune cells, recognizes unmethylated CpG oligonucleodides and is responsible for innate immune cells activation and mounting acquired responses. The NF‐κB luciferase assay showed that p.E317D, located within the ligand sensor pocket of TLR9, caused a 50% reduction in receptor activation. TLR9 resequencing identified additional rare variants carriers: 7 in EOAD (7/320, 2.19%) and 19 in LOAD (19/1166, 1.63%) patients and 8 in controls (8/934, 0.86%), while TLR9 p.E317D was absent. Of the rare variants, 16 were in patient‐only, 3 in patients and controls and 5 in controls‐only. The NF‐κB luciferase assay indicated reduced TLR9 activation for the novel variants specific to EOAD patients and with a CADD score >20.ConclusionWe discovered TLR9 as candidate gene in AD with a mechanism of loss of receptor activation.

Non-random distribution of deleterious mutations in the DNA and protein-binding domains of IRF6 are associated with Van Der Woude syndrome.

BackgroundThe development of the face occurs during the early days of intrauterine life by the formation of facial processes from the first Pharyngeal arch. Derangement in these well‐organized fusion events results in Orofacial clefts (OFC). Van der Woude syndrome (VWS) is one of the most common causes of syndromic cleft lip and/or palate accounting for 2% of all cases. Mutations in the IRF6 gene account for 70% of cases with the majority of these mutations located in the DNA‐binding (exon 3, 4) or protein‐binding domains (exon 7–9). The current study was designed to update the list of IRF6 variants reported for VWS by compiling all the published mutations from 2013 to date as well as including the previously unreported VWS cases from Africa and Puerto Rico.MethodsWe used PubMed with the search terms; "Van der Woude syndrome," “Popliteal pterygium syndrome,” "IRF6," and "Orofacial cleft" to identify eligible studies. We compiled the CADD score for all the mutations to determine the percentage of deleterious variants.ResultsTwenty‐one new mutations were identified from nine papers. The majority of these mutations were in exon 4. Mutations in exon 3 and 4 had CADD scores between 20 and 30 and mutations in exon 7–9 had CADD scores between 30 and 40. The presence of higher CADD scores in the protein‐binding domain (exon 7–9) further confirms the crucial role played by this domain in the function of IRF6. In the new cases, we identified five IRF6 mutations, three novel missense mutations (p.Phe36Tyr, p.Lys109Thr, and p.Gln438Leu), and two previously reported nonsense mutations (p.Ser424*and p.Arg250*).ConclusionMutations in the protein and DNA‐binding domains of IRF6 ranked among the top 0.1% and 1% most deleterious genetic mutations, respectively. Overall, these findings expand the range of VWS mutations and are important for diagnostic and counseling purposes.

An Integrated Deep-Mutational-Scanning Approach Provides Clinical Insights on PTEN Genotype-Phenotype Relationships

Germline variation in PTEN results in variable clinical presentations, including benign and malignant neoplasia and neurodevelopmental disorders. Despite decades of research, it remains unclear how the PTEN genotype is related to clinical outcomes. In this study, we combined two recent deep mutational scanning (DMS) datasets probing the effects of single amino acid variation on enzyme activity and steady-state cellular abundance with a large, well-curated clinical cohort of PTEN-variant carriers. We sought to connect variant-specific molecular phenotypes to the clinical outcomes of individuals with PTEN variants. We found that DMS data partially explain quantitative clinical traits, including head circumference and Cleveland Clinic (CC) score, which is a semiquantitative surrogate of disease burden. We built logistic regression models that use DMS and CADD scores to separate clinical PTEN variation from gnomAD control-only variation with high accuracy. By using a survival-like analysis, we identified molecular phenotype groups with differential risk of early cancer onset as well as lifetime risk of cancer. Finally, we identified classes of DMS-defined variants with significantly different risk levels for classical hamartoma-related features (odds ratio [OR] range of 4.1-102.9). In stark contrast, the risk for developing autism or developmental delay does not significantly change across variant classes (OR range of 5.4-12.4). Together, these findings highlight the potential impact of combining DMS datasets with rich clinical data and provide new insights that might guide personalized clinical decisions for PTEN-variant carriers.

VarFish: comprehensive DNA variant analysis for diagnostics and research.

VarFish is a user-friendly web application for the quality control, filtering, prioritization, analysis, and user-based annotation of DNA variant data with a focus on rare disease genetics. It is capable of processing variant call files with single or multiple samples. The variants are automatically annotated with population frequencies, molecular impact, and presence in databases such as ClinVar. Further, it provides support for pathogenicity scores including CADD, MutationTaster, and phenotypic similarity scores. Users can filter variants based on these annotations and presumed inheritance pattern and sort the results by these scores. Variants passing the filter are listed with their annotations and many useful link-outs to genome browsers, other gene/variant data portals, and external tools for variant assessment. VarFish allows users to create their own annotations including support for variant assessment following ACMG-AMP guidelines. In close collaboration with medical practitioners, VarFish was designed for variant analysis and prioritization in diagnostic and research settings as described in the software's extensive manual. The user interface has been optimized for supporting these protocols. Users can install VarFish on their own in-house servers where it provides additional lab notebook features for collaborative analysis and allows re-analysis of cases, e.g. after update of genotype or phenotype databases.

Germline GAB2 Mutations in Childhood Acute Lymphoblastic Leukemia

Recent studies using next-generation sequencing of selected individuals, such as those with familial leukemia or congenital syndromes, have identified rare and highly penetrant germline mutations that predispose to childhood acute lymphoblastic leukemia (ALL). High hyperdiploidy (HD), the most common cytogenetic subtype of childhood ALL, is enriched in children with RASopathies who develop ALL and, similarly, a high proportion of ALL patients with germline ETV6 or IKZF1 mutations presented with the HD subtype. Here, we aimed to identify novel predisposition genes in a selected group of HD-ALL patients.Targeted sequencing of 538 cancer-relevant genes was carried out using the UCSF500 Cancer Gene Panel in diagnostic bone marrow (i.e. tumor) DNA from 57 HD-ALL patients from the California Childhood Leukemia Study (CCLS). HD-ALL patients were selected based on absence of somatic KRAS or NRAS hotspot mutations detectable by Sanger sequencing, and absence of somatic copy number deletions from multiplex ligation-dependent probe amplification (MLPA) assays. After filtering out likely somatic mutations (mutant allele fraction <0.44), and restricting to variants with low frequency in unselected individuals (allele frequency <0.01% in the Exome Aggregation Consortium, ExAC) and with predicted functional effects (Combined Annotation Dependent Depletion, CADD score ≥20), we identified 151 putative predisposing mutations. Of 41 mutations of interest selected for validation, 37 (90.2%) were confirmed as germline in origin via Sanger sequencing of remission or newborn bloodspot DNA.Rare and predicted functional germline mutations in known (NBN, SH2B3, ETV6, CREBBP, MSH6) or suspected (MLL, ABL1, FLT3, MYH9) ALL predisposition genes were identified in nine out of 57 patients (15.8%). Three additional patients harbored germline mutations in the GRB2-associated binding protein 2 (GAB2), a known binding partner of PTPN11-encoded SHP2 and activator of the ERK/MAPK and PI3K/AKT pathways. Two GAB2 mutations, a missense mutation S592F and frameshift mutation P621fs, were predicted to be highly functional (CADD scores = 34 and 36 respectively) and absent in ExAC. Frequency of rare and damaging GAB2 mutations was significantly higher in our patient set (2.6%) than in ExAC (0.28%, P = 2.70 x 10-6). We replicated this finding in sequencing data from 309 ALL patients in the TARGET (Therapeutically Applicable Research to Generate Effective Treatments) project (0.81% vs. 0.28%, P = 0.015).Patient GAB2 mutations were cloned into HEK293 cells and, following EGF stimulation, we found that the P621fs mutation reduced SHP2 binding and ERK1/2 phosphorylation but increased AKT phosphorylation. This suggested possible Ras-independent leukemogenic effects, supported by a lack of somatic Ras pathway mutations in the three GAB2 mutant patients. Additional functional analyses and sequencing of larger patient cohorts will be required to elucidate the role of germline GAB2 mutations in childhood ALL. DisclosuresNo relevant conflicts of interest to declare.

Whole exome sequencing identified sixty-five coding mutations in four neuroblastoma tumors

Neuroblastoma is a pediatric tumor characterized by histologic heterogeneity, and accounts for ~15% of childhood deaths from cancer. The five-year survival for patients with high-risk stage 4 disease has not improved in two decades. We used whole exome sequencing (WES) to identify mutations present in three independent high-risk stage 4 neuroblastoma tumors (COA/UAB-3, COA/UAB -6 and COA/UAB -8) and a stage 3 tumor (COA/UAB-14). Among the four tumors WES analysis identified forty-three mutations that had not been reported previously, one of which was present in two of the four tumors. WES analysis also corroborated twenty-two mutations that were reported previously. No single mutation occurred in all four tumors or in all stage 4 tumors. Three of the four tumors harbored genes with CADD scores ≥20, indicative of mutations associated with human pathologies. The average depth of coverage ranged from 39.68 to 90.27, with >99% sequences mapping to the genome. In summary, WES identified sixty-five coding mutations including forty-three mutations not reported previously in primary neuroblastoma tumors. The three stage 4 tumors contained mutations in genes encoding protein products that regulate immune function or cell adhesion and tumor cell metastasis.

INCREASED BURDEN OF RARE LOSS-OF-FUNCTION VARIANTS IN ALZHEIMER’S DISEASE PATIENTS COMPARED TO CENTENARIANS

Late onset Alzheimer's disease (AD) has high estimated heritability, but the effects of common genetic variants could explain only 24% to 33% of total phenotypic variance for AD. In the present study we investigated the burden of rare loss-of-function variants in a whole genome sequencing study of 45 AD subjects and 53 cognitively healthy centenarians, all of Ashkenazi Jewish ancestry (AJ). We restricted our analysis to coding rare variants observed at most once in the 1000 Genomes database and having a minor allele frequency below 2% in our AJ sample. We compared the burden of predicted protein altering variants between cases and controls as normalized by the level of rare synonymous variants. We observed an increased burden among AD subjects for predicted loss-of-function variants (LoFs) defined as stop-gain, frame shift, initiation codon (INIT), and splice site mutations (n=930, OR=1.3, P=1.5xE-5). Among LoFs, the strongest burden was observed for INIT (OR=2.16, P=0.0097) and premature stop variants predicted to cause nonsense mediated decay in the majority of transcripts (NMD) (OR=1.98, P=0.02). Notably, this increased burden of NMD, INIT and splice variants was more pronounced in innate immune genes (Gene Ontology ID: 0045087) (OR=4.55, P=0.0043). Further comparison to additional whole exomes of age-matched AJ samples indicates that the difference in LoF burden originated both from the AD and centenarian sample. To exploration how non-coding variants differ between AD and centenarians, we obtained Combined Annotation Dependent Depletion CADD scores as a measure of deleteriousness for rare non-coding variants. In non-coding regions, the enrichment of potentially functional rare variants with the highest CADD scores (Phred>=40) compared to those with the lowest (Phred <=0.2) in AD patients and centenarians was not significant, but in the same direction (OR=1.3, P=0.21) as the LoFs in AD. In summary, we observed an overall increased burden of rare Loss-of-function variants in AD subjects as compared to centenarians, and this enrichment is more pronounced for innate immune genes. Replication of the coding region enrichment in a larger sample and investigation of non-coding functional variants are on ongoing.

Abstract 2704: Pathogenic mutations and variants of unknown significance (VUS) in cancer predisposition genes are associated with over 10% of pediatric rhabdomyosarcoma: a report from the Children’s Oncology Group

Abstract Approximately 5% of rhabdomyosarcoma (RMS) cases are due to known cancer predisposition syndromes (e.g., Li-Fraumeni syndrome, Neurofibromatosis-1), but these estimates have not been confirmed in large-scale studies. Furthermore, no recommended germline testing protocols exist for RMS. We tested the hypothesis that germline mutation burden is greater than previously reported and present several new predisposition genes as potential drivers of pediatric RMS. We sequenced 59 cancer susceptibility genes in 213 children with RMS enrolled on one clinical trial (COG ARST0531), unselected for family history of cancer. Validation was performed with Sanger sequencing. Our analysis included determining the incidence of pathogenic variants in known RMS predisposition genes, followed by other cancer predisposition genes on our panel but not previously associated with RMS. In our cohort, 3.9% of unselected RMS cases harbor a pathogenic variant in a known predisposition gene (Table 1). Using VAAST, we next identified the top 10 genes possibly associated with RMS. Variants of uncertain significance (VUS) were restricted to rare variants (&amp;lt;0.1%) with severe CADD and Polyphen2 scores, which added 16 additional patients (7.4%). To distinguish incidental findings from true correlations, we compared our allele frequencies to those in the GnomAD database (126,216 exome sequences and 15,136 whole-genome sequences). Including both pathogenic mutations and rare VUS, 10.8% of RMS cases carry a variant with potentially important clinical implications, suggesting that newly diagnosed RMS would benefit from multigene genetic testing. Future directions include linkage to clinical data such as family history, age at diagnosis, tumor stage/location/histology, and clinical outcome, as well as expanding both the RMS and control cohorts. Table 1.List of genes with numbers of carriers identified in 213 rhabdomyosarcoma casesGeneCarriers%% of carriers in GnomADSIR95% CIP=valueTruncating Variants and Known Pathogenic MissenseATM00.42%---BAP100.87%0.05%---BRCA220.87%0.85%1.01(0.12-3.66)1.1736DICER110.43%0.06%7.75(0.20-43.19)0.2421MSH210.43%0.15%2.93(0.07-16.30)0.5790NF120.87%0.88%0.98(0.12-3.54)1.2088PMS200.87%0.18%---PTCH100.15%---TP5331.30%0.13%10.10(2.08-29.51)0.0070WT101.30%0.02%---Total93.90%3.42%1.14(0.52-2.16)0.7882Combined Analysis of Pathogenic Variants plus Variants of Uncertain Significance (VUS)ATM20.87%0.78%1.11(0.13-4.02)1.0734BAP120.87%0.22%3.89(0.47-14.05)0.1891BRCA220.87%1.52%0.57(0.07-2.06)1.7286DICER173.03%0.86%3.52(1.41-7.24)0.0089MSH210.43%0.78%0.56(0.01-3.10)1.6690NF120.87%1.70%0.51(0.06-1.84)1.8062PMS210.43%0.64%0.68(0.02-3.78)1.5425PTCH131.30%0.86%1.52(0.31-4.43)0.6348TP5341.73%0.25%6.93(1.89-17.76)0.0058WT110.43%0.31%1.40(0.04-7.81)1.0197Total2510.82%9.94%1.09(0.70-1.61)0.7232 Citation Format: Erin L. Young, Luke Maese, Rosann Robinson, Lance Pflieger, Barry Moore, Shawn Rynearson, Trent Fowler, Sean V. Tavtigian, Mark Yandell, Clinton C. Mason, Douglas S. Hawkins, Philip J. Lupo, Joshua D. Schiffman. Pathogenic mutations and variants of unknown significance (VUS) in cancer predisposition genes are associated with over 10% of pediatric rhabdomyosarcoma: a report from the Children’s Oncology Group [abstract]. In: Proceedings of the American Association for Cancer Research Annual Meeting 2017; 2017 Apr 1-5; Washington, DC. Philadelphia (PA): AACR; Cancer Res 2017;77(13 Suppl):Abstract nr 2704. doi:10.1158/1538-7445.AM2017-2704

WHOLE-GENOME SEQUENCING IN FAMILIAL LATE-ONSET ALZHEIMER’S DISEASE IDENTIFIES RARE VARIATION IN AD CANDIDATE GENES

The Alzheimer's Disease Sequencing Project (ADSP) is an initiative to identify rare genetic variation influencing Late Onset Alzheimer's Disease (LOAD) risk. As part of the ADSP, we performed whole-genome sequencing (WGS) in 67 Caribbean Hispanic (CH) and 44 non-Hispanic white (NHW) extended families multiply affected by LOAD, followed by extensive quality control, variant filtering, and gene-based association tests. WGS data were generated for 351 subjects from 67 CH families and 197 subjects from 44 NHW families, including both AD and cognitively intact relatives. Alignment was performed using the Burrows-Wheeler algorithm, followed by genotype calling using a consensus calling pipeline that used both GATK genotype calls and ATLAS genotype calls. Variants were annotated for allele frequency and predicted functional impact, categorized into damaging (e.g., loss of function, high CADD scores, etc) and likely damaging (e.g., non-synonymous, moderate CADD, etc). We performed gene-based association testing, accounting for family structure using the FSKAT software. Association was performed using rare variants (MAF<0.01) and two models (damaging set only; damaging and likely damaging), with CH and NHW sets analyzed separately. Examination of the 30 known LOAD genes (largely identified by GWAS-based meta-analyses) confirmed the role of rare functional variation in a number of genes, including CR1 (p =0.040 in NHW, p=0.049 in CH; damaging variants only) and SLC24A4 (p=0.002 in NHW, p=0.040 in CH). Other candidate genes showed nominal association in one but not both datasets (AKAP9, FERMT2, GRN, HLA-DRB5, MEF2C in NHW dataset; BIN1, PTK2B in CH). Additional genes showed strong association in the CH dataset, with nominal association in the NHW. These include ATG2A (p=0.0003 in CH, p=0.02 in NHW), a gene involved in both autophagy and lipid droplet formation, and CD69 (p=0.00054 in CH, p=0.02 in NHW), a gene involved in cell proliferation. These results suggest rare, functional variation may influence LOAD risk in multiplex families, even among genes identified through common variation. We also show association with novel genes, ATG2A and CD69, with support from both CH and NHW families. Variants are currently being validated using other technologies, and follow-up and replication analyses are ongoing.

Identification of putative second genetic hits in schizophrenia carriers of high-risk copy number variants and resequencing in additional samples

Copy number variants (CNVs) conferring risk of schizophrenia present incomplete penetrance, suggesting the existence of second genetic hits. Identification of second hits may help to find genes with rare variants of susceptibility to schizophrenia. The aim of this work was to search for second hits of moderate/high risk in schizophrenia carriers of risk CNVs and resequencing of the relevant genes in additional samples. To this end, ten patients with risk CNVs at cytobands 15q11.2, 15q11.2-13.1, 16p11.2, or 16p13.11, were subjected to whole-exome sequencing. Rare single nucleotide variants, defined as those absent from main public databases, were classified according to bioinformatic prediction of pathogenicity by CADD scores. The average number of rare predicted pathogenic variants per sample was 13.6 (SD 2.01). Two genes, BFAR and SYNJ1, presented rare predicted pathogenic variants in more than one sample. Follow-up resequencing of these genes in 432 additional cases and 432 controls identified a significant excess of rare predicted pathogenic variants in case samples at SYNJ1. Taking into account its function in clathrin-mediated synaptic vesicle endocytosis at presynaptic terminals, our results suggest an impairment of this process in schizophrenia.

GAVIN: Gene-Aware Variant INterpretation for medical sequencing

We present Gene-Aware Variant INterpretation (GAVIN), a new method that accurately classifies variants for clinical diagnostic purposes. Classifications are based on gene-specific calibrations of allele frequencies from the ExAC database, likely variant impact using SnpEff, and estimated deleteriousness based on CADD scores for >3000 genes. In a benchmark on 18 clinical gene sets, we achieve a sensitivity of 91.4% and a specificity of 76.9%. This accuracy is unmatched by 12 other tools. We provide GAVIN as an online MOLGENIS service to annotate VCF files and as an open source executable for use in bioinformatic pipelines. It can be found at http://molgenis.org/gavin.

T2. IDENTIFICATION OF PATHOGENIC VARIANTS IN PROTEIN CODING GENES

Background A central aim of precision medicine is to target treatments to the underlying causes of disease. To accurately target treatments we must be able to recognize pathogenic genetic variants. Current methods prioritize variants that directly alter protein sequence (missense and loss of function) but not variants that may cause disease by changing the processing of final transcripts. The difficulty in capturing this effect results in overlooking synonymous and intronic variants when searching for disease risk in sequenced genomes. Methods The TraP score was constructed using three main components: 1) Information acquisition – details of the harboring gene and it’s transcripts are gathered for each variant. 2) Feature calculation - possible changes to sequence motifs are evaluated, including changes to exon-intron boundaries, creation of cryptic splice sites, creations and disruptions of cis-acting binding sites for splicing regulatory proteins, interactions between selected features such as original and new splice sites and others. Overall, 42 features and 14 general properties (chromosome, strand, coordinate, etc.) are collected for each variant. 3) Modeling – the incorporation of selected features into a random forest model. The model is trained on a set of 75 pathogenic synonymous variants and 402 benign variants. Pathogenic variants are strongly associated with rare disease, whereas the 402 benign variants are de novo mutations identified from healthy individuals. Results The Transcript-inferred Pathogenicity score (TraP) presented here was constructed to reliably identify non-coding mutations that cause disease. Trap is strongly negatively correlated with allele frequency in both synonymous and intronic regions, suggesting that the higher the TraP score the stronger the selection against these variants in the population. Moreover, synonymous variants with high TraP scores have significantly lower minor allele frequencies than even missense variants, indicating that Trap identifies a subset of synonymous variants under stronger purifying selection. TraP identifies known pathogenic variants in synonymous and intronic ClinVar datasets (AUC = 0.88 and 0.83, respectively), dismissing benign variants with extremely high specificity of above 99%. Applied to exomes of 281 epilepsy family trios, TraP pinpoints synonymous de novo variants in known epilepsy genes. TraP’s high performance and specificity clearly outperforms existing methods and allows the prioritization of synonymous and intronic variants for use in gene discovery and the interpretation of personal genomes. Discussion Exome sequencing studies consider rare non-synonymous variants as disease candidates, while other variant types are mostly ignored. Some existing methods are able to prioritize synonymous and intronic variants, yet lack the specificity required for detection of causal variants. TraP discards over 99% of non-coding variants as benign while strongly identifying true pathogenic variants. TraP identifies pathogenic variants that are not conserved, yet have rare population frequencies. Doing so without prior population frequency information and in contrast to the GERP++ and CADD scores, suggests that TraP identifies pathogenic events that were not selected against during vertebrate evolution, but are selected against in human population. This conclusion is supported by the highest complexity of alternative splicing found in primates and by the species-specific nature of splicing regulation.

Abstract 12066: TIMP3 and CHADL Mutations Are Common in Turner Syndrome Aortopathy

Introduction: Turner syndrome (TS) is caused by the loss of the second sex chromosome, which occurs in approximately 1 in 2,000 female births. TS alone is an independent risk factor for aortic enlargement. 25 to 50% of those with TS also have a bicuspid aortic valve (BAV) which is present in most who have aortic dissection and rupture. Methods: We used whole exome sequencing to study the genes of 188 TS subjects with BAV (88 cases) or normal aortic valves (100 controls). Aorta diameters as markers of aortopathy severity converted into Z-scores were utilized as covariates with BAV status, and as a sole predictor of genetic variation. Enlarged aorta was defined as an ascending aorta Z-score &gt; 2. Statistical analyses employed a kernel gene-based method, SKAT-O. Results: TS girls and women with BAV had significantly higher ascending aortic Z-scores than those without BAV (1.95 ± 1.97 vs 0.55 ± 1.01, t-test p= 1.6 x 10 -5 ).Mutations in TIMP3 and CHADL achieved exome-wide significance for association with BAV/aortic dilatation (p=1.6x10 -6 and 1.1x10 -6 , respectively). TIMP3 significance increased when aortic root Z-scores were included as a continuous covariant with BAV status (p=2.3x10 -7 ). Overall, 30.7% of those with aortic dilatation had mutations predicted by CADD scores to be deleterious in TIMP3 (8 individuals), CHADL (4 individuals), or both (4 individuals). Those with mutations in both genes were more likely to have enlarged aortas than those with 1 or none (chi-squared p=2.8x10 -3 ). Discussion and conclusions: BAV is known to be associated with aortopathy in TS. We found that TIMP3 and CHADL mutations occur commonly with those findings . TIMP3 is an inhibitor of several aortic matrix metalloproteinases, a group of peptidases that degrade extracellular matrix (ECM). CHADL is a collagen binding protein that regulates collagen fiber formation. Collagens are necessary for aortic structural integrity. We hypothesize that mutations in TIMP3 and CHADL disrupt the critical balance between ECM creation and degradation, thereby compromising aortic strength. There appears to be a synergy between TIMP3 and CHADL functions that may be involved in both BAV development and aortopathy. Both TIMP3 and CHADL may serve as biomarkers and pharmacologic targets for aortic disease in TS.

OR41 A survey of cancer somatic mutations in the HLA region

Although a number of cancer associations with HLA region SNPs have been reported, there is still no conclusion on whether the HLA complex contains cancer susceptibility genes or if the associations reflect alterations in immune surveillance. To identify the HLA-linked non-HLA genes playing a role in carcinogenesis, we screened germline variants of the HLA region (chr6:28.5 to 34.5 Mb; hg19) that also exist as somatic mutations in cancers. Examination of 285,609 HLA region SNPs in the COSMIC database revealed that 660 of them have been detected in 747 cancer samples, some in up to three different cancers. Most ( n = 248) of the 660 SNPs detected as cancer somatic mutations were missense SNPs, with 36 located in the HLA-A, B, C, DRA/DRB1, or DQA1/DQB1 genes, and only few were assessed as possibly damaging by PolyPhen. Overall, 28.6% of the SNPs were highly deleterious (in the top one percentile in the genome), and likely to be driver mutations as assessed by CADD scores (>20; overall median = 13.7, as compared to median CADD score of 7.0 for missense mutations). The HLA region genes most deleteriously mutated in cancer consisted of antigen processing pathway genes TAP1/2/BP and PSMB8/9, and cancer-related genes MSH5 and DDR1. Of these, TAPBP expression showed a correlation with breast and gastric cancer survival in the Kaplan–Meier Plotter ( P P