Duodenal Fluid Analysis as a Rewarding Approach to Detect Low-Abundance Mutations in Biliopancreatic Cancers.

Diagnosis of biliopancreatic cancers by the available serum tumor markers, imaging, and histopathological tissue specimen examination remains a challenge. Circulating cell-free DNA derived from matched pairs of secretin-stimulated duodenal fluid (DF) and plasma from 10 patients with biliopancreatic diseases and 8 control subjects was analyzed using AmpliSeq™ HD technology for Ion Torrent Next-Generation Sequencing to evaluate the potential of liquid biopsy with DF in biliopancreatic cancers. The median cfDNA concentration was greater in DF-derived than in plasma-derived samples. A total of 13 variants were detected: 11 vs. 1 were exclusive for DF relative to the plasma source, and 1 was shared between the two body fluids. According to the four-tier systems, 10 clinical tier-I-II (76.9%), 1 tier-III (7.7%), and 2 tier-IV (15.4%) variants were identified. Notably, the 11 tier-I-III variants were exclusively found in DF-derived cfDNA from five patients with biliopancreatic cancers, and were detected in seven genes (KRAS, TP53, BRAF, CDKN2A, RNF43, GNAS, and PIK3CA); 82% of the tier-I-III variants had a low abundance, with a VAF < 6%. The mutational profiling of DF seems to be a reliable and promising tool for identifying cancer-associated alterations in malignant cancers of the biliopancreatic tract.

Paired comparison of the analytical performance between the Oncomine™ Comprehensive Assay v3 and whole-exome sequencing of ovarian cancer tissue.

BACKGROUND: Next-generation sequencing (NGS) has been implemented in clinical oncology as a personalized medicine tool to identify targetable genetic alterations and to guide treatment decisions. However, the optimal NGS test strategy and target genes for clinical use are still being discussed. The aim was to compare the performance of the Oncomine™ Comprehensive Assay v3 (OCAv3) (targeted gene panel) and whole-exome sequencing (WES) to investigate somatic single and multiple nucleotide variants and small indels in ovarian cancer patients. METHODS AND RESULTS: Genomic DNA was isolated from fresh frozen samples of five high-grade serous (HGSC) and three clear cell ovarian (oCCC) cancer patients. Exome sequencing libraries were prepared by using the Ion AmpliSeq Exome RDY kit, whereas libraries for OCAv3 were prepared using by Ion AmpliSeq™ Library Kit Plus. Sequencing was performed using the Ion S5XL System (Thermo Fisher Scientific). When including only variants classified as pathogenic, likely pathogenic or unknown significance based on ClinVar database verdicts and comparing overlapping regions covered both by the OCAv3 assay and WES, 23 variants were detected by both assays. However, OCAv3 detected additionally two variants: ARID1A: p.Gln563Ter and TP53: p.Ser261ValfsTer84 that have not passed WES filtering criteria due to low coverage. CONCLUSIONS: With the present treatment possibilities, OCAv3 panel testing provided higher diagnostic yield due to better coverage. Our study emphasizes that WES, although offering the potential to identify novel findings in genes not covered by OCAv3, might overlook variants in genes relevant for OC.

High-confidence calling of normal epithelial cells allows identification of a novel stem-like cell state in the colorectal cancer microenvironment.

Single-cell analyses can be confounded by assigning unrelated groups of cells to common developmental trajectories. For instance, cancer cells and admixed normal epithelial cells could adopt similar cell states thus complicating analyses of their developmental potential. Here, we develop and benchmark CCISM (for Cancer Cell Identification using Somatic Mutations) to exploit genomic single nucleotide variants for the disambiguation of cancer cells from genomically normal non-cancer cells in single-cell data. We find that our method and others based on gene expression or allelic imbalances identify overlapping sets of colorectal cancer versus normal colon epithelial cells, depending on molecular characteristics of individual cancers. Further, we define consensus cell identities of normal and cancer epithelial cells with higher transcriptome cluster homogeneity than those derived using existing tools. Using the consensus identities, we identify significant shifts of cell state distributions in genomically normal epithelial cells developing in the cancer microenvironment, with immature states increased at the expense of terminal differentiation throughout the colon, and a novel stem-like cell state arising in the left colon. Trajectory analyses show that the new cell state extends the pseudo-time range of normal colon stem-like cells in a cancer context. We identify cancer-associated fibroblasts as sources of WNT and BMP ligands potentially contributing to increased plasticity of stem cells in the cancer microenvironment. Our analyses advocate careful interpretation of cell heterogeneity and plasticity in the cancer context and the consideration of genomic information in addition to gene expression data when possible.

Understanding variants of unknown significance: the computational frontier.

The rapid advancement of sequencing technologies has led to the identification of numerous mutations in cancer genomes, many of which are variants of unknown significance (VUS). Computational models are increasingly being used to predict the functional impact of these mutations, in both coding and noncoding regions. Integration of these models with emerging genomic datasets will refine our understanding of mutation effects and guide clinical decision making. Future advancements in modeling protein interactions and transcriptional regulation will further enhance our ability to interpret VUS. Periodic incorporation of these developments into VUS reclassification practice has the potential to significantly improve personalized cancer care.

Mutation spectra of the BRCA1/2 genes in human breast and ovarian cancer and germline.

Annotating genomic sequence alterations is sometimes a difficult decision, particularly in missense variants with uncertain pathogenic significance and also in those presumed as germline pathogenic variants. We here suggest that mutation spectrum may also be useful for judging them. From the public databases, 982 BRCA1/1861 BRCA2 germline missense variants and 294 BRCA1/420 BRCA2 somatic missense variants were obtained. We then compared their mutation spectra, i.e., the frequencies of two transition- and four transversion-type mutations, in each category. Intriguingly, in BRCA1 variants, A:T to C:G transversion, which was relatively frequent in the germline, was extremely rare in somatic, particularly breast cancer, cells (p = .03). Conversely, A:T to T:A transversion was most infrequent in the germline, but not rare in somatic cells. Thus, BRCA1 variants with A:T to T:A transversion may be suspected as somatic, and those with A:T to C:G as being in the germline. These tendencies of mutation spectrum may also suggest the biological and chemical origins of the base alterations. On the other hand, unfortunately, variants of uncertain significance (VUS) were not distinguishable by mutation spectrum. Our findings warrant further and more detailed studies.

Genetic alterations in mature B- and T-cell lymphomas - a practical guide to WHO-HAEM5.

The identification of recurrent genomic alterations in tumour cells has a significant role in the classification of mature B- and T-cell lymphomas. Following the development of new technologies, such as next generation sequencing and the improvement of classical technologies such as conventional and molecular cytogenetics, a huge catalogue of genomic alterations in lymphoid neoplasms has been established. These alterations are relevant to refine the taxonomy of the classification of lymphomas, to scrutinize the differential diagnosis within different lymphoma entities and to help assessing the prognosis and clinical management of the patients. Consequently, here we describe the key genetic alterations relevant in mature B- and T-cell lymphomas.

Somatic mutational profiling and clinical impact of driver genes in Latin-Iberian medulloblastomas: Towards precision medicine.

Medulloblastoma (MB) is the most prevalent malignant brain tumor in children, known for its heterogeneity and treatment-associated toxicity, and there is a critical need for new therapeutic targets. We analyzed the somatic mutation profile of 15 driver genes in 69 Latin-Iberian molecularly characterized medulloblastomas using the Illumina TruSight Tumor 15 panel. We classified the variants based on their clinical impact and oncogenicity. Among the patients, 66.7% were MBSHH, 13.0% MBWNT, 7.3% MBGrp3, and 13.0% MBGrp4. Among the 63 variants found, 54% were classified as Tier I/II and 31.7% as oncogenic/likely oncogenic. We observed 33.3% of cases harboring at least one mutation. TP53 (23.2%, 16/69) was the most mutated gene, followed by PIK3CA (5.8%, 4/69), KIT (4.3%, 3/69), PDGFRA (2.9%, 2/69), EGFR (1.4%, 1/69), ERBB2 (1.4%, 1/69), and NRAS (1.4%, 1/69). Approximately 41% of MBSHH tumors exhibited mutations, TP53 (32.6%) being the most frequently mutated gene. Tier I/II and oncogenic/likely oncogenic TP53 variants were associated with relapse, progression, and lower survival rates. Potentially actionable variants in the PIK3CA and KIT genes were identified. Latin-Iberian medulloblastomas, particularly the MBSHH, exhibit higher mutation frequencies than other populations. We corroborate the TP53 mutation status as an important prognostic factor, while PIK3CA and KIT are potential therapeutic targets.

Topical hidden genome: discovering latent cancer mutational topics using a Bayesian multilevel context-learning approach.

Inferring the cancer-type specificities of ultra-rare, genome-wide somatic mutations is an open problem. Traditional statistical methods cannot handle such data due to their ultra-high dimensionality and extreme data sparsity. To harness information in rare mutations, we have recently proposed a formal multilevel multilogistic "hidden genome" model. Through its hierarchical layers, the model condenses information in ultra-rare mutations through meta-features embodying mutation contexts to characterize cancer types. Consistent, scalable point estimation of the model can incorporate 10s of millions of variants across thousands of tumors and permit impressive prediction and attribution. However, principled statistical inference is infeasible due to the volume, correlation, and noninterpretability of mutation contexts. In this paper, we propose a novel framework that leverages topic models from computational linguistics to effectuate dimension reduction of mutation contexts producing interpretable, decorrelated meta-feature topics. We propose an efficient MCMC algorithm for implementation that permits rigorous full Bayesian inference at a scale that is orders of magnitude beyond the capability of existing out-of-the-box inferential high-dimensional multi-class regression methods and software. Applying our model to the Pan Cancer Analysis of Whole Genomes dataset reveals interesting biological insights including somatic mutational topics associated with UV exposure in skin cancer, aging in colorectal cancer, and strong influence of epigenome organization in liver cancer. Under cross-validation, our model demonstrates highly competitive predictive performance against blackbox methods of random forest and deep learning.

Germline potential should not be overlooked for cancer variants identified in tumour-only somatic mutation testing.

DNA sequencing of tumour tissue has become the standard care for many solid cancers because of the option to detect somatic variants that have significant therapeutic, diagnostic and prognostic implications. Variants found within the tumour may be either somatic or germline in origin. Somatic cancer gene panels are developed to detect acquired (somatic) variants that are relevant for therapeutic or molecular characterisation of the tumour, expanding gene panels now include genes which may also inform patient management such as cancer predisposition syndromes (CPS) genes. Identifying germline cancer predisposition variants can alter cancer management, the risk of developing new primary cancers and risk for cancer in at-risk family members. This paper discusses the clinical, technical and ethical challenges related to identifying and reporting potential germline pathogenic variants that are detected on tumour sequencing. It also highlights the existence of the eviQ national guidelines for CPS with advice on germline confirmation of somatic findings to pathology laboratories in Australia.

Somatic variants as a cause of drug-resistant epilepsy including mesial temporal lobe epilepsy with hippocampal sclerosis.

OBJECTIVE: The contribution of somatic variants to epilepsy has recently been demonstrated, particularly in the etiology of malformations of cortical development. The aim of this study was to determine the diagnostic yield of somatic variants in genes that have been previously associated with a somatic or germline epilepsy model, ascertained from resected brain tissue from patients with multidrug-resistant focal epilepsy. METHODS: Forty-two patients were recruited across three categories: (1) malformations of cortical development, (2) mesial temporal lobe epilepsy with hippocampal sclerosis, and (3) nonlesional focal epilepsy. Participants were subdivided based on histopathology of the resected brain. Paired blood- and brain-derived DNA samples were sequenced using high-coverage targeted next generation sequencing to high depth (585× and 1360×, respectively). Variants were identified using Genome Analysis ToolKit (GATK4) MuTect-2 and confirmed using high-coverage Amplicon-EZ sequencing. RESULTS: Sequence data on 41 patients passed quality control. Four somatic variants were validated following amplicon sequencing: within CBL, ALG13, MTOR, and FLNA. The diagnostic yield across 41 patients was 10%, 9% in mesial temporal lobe epilepsy with hippocampal sclerosis and 20% in malformations of cortical development. SIGNIFICANCE: This study provides novel insights into the etiology of mesial temporal lobe epilepsy with hippocampal sclerosis, highlighting a potential pathogenic role of somatic variants in CBL and ALG13. We also report candidate diagnostic somatic variants in FLNA in focal cortical dysplasia, while providing further insight into the importance of MTOR and related genes in focal cortical dysplasia. This work demonstrates the potential molecular diagnostic value of variants in both germline and somatic epilepsy genes.

Enrichment of Immune Dysregulation Disorders in Adult Patients with Human Inborn Errors of Immunity.

Human inborn errors of immunity (IEI) comprise a group of diseases resulting from molecular variants that compromise innate and adaptive immunity. Clinical features of IEI patients are dominated by susceptibility to a spectrum of infectious diseases, as well as autoimmune, autoinflammatory, allergic, and malignant phenotypes that usually appear in childhood, which is when the diagnosis is typically made. However, some IEI patients are identified in adulthood due to symptomatic delay of the disease or other reasons that prevent the request for a molecular study. The application of next-generation sequencing (NGS) as a diagnostic technique has given rise to an ever-increasing identification of IEI-monogenic causes, thus improving the diagnostic yield and facilitating the possibility of personalized treatment. This work was a retrospective study of 173 adults with IEI suspicion that were sequenced between 2005 and 2023. Sanger, targeted gene-panel, and whole exome sequencing were used for molecular diagnosis. Disease-causing variants were identified in 44 of 173 (25.43%) patients. The clinical phenotype of these 44 patients was mostly related to infection susceptibility (63.64%). An enrichment of immune dysregulation diseases was found when cohorts with molecular diagnosis were compared to those without. Immune dysregulation disorders, group 4 from the International Union of Immunological Societies Expert Committee (IUIS), were the most prevalent among these adult patients. Immune dysregulation as a new item in the Jeffrey Model Foundation warning signs for adults significantly increases the sensitivity for the identification of patients with an IEI-producing molecular defect.

Whole-exome sequencing of Nigerian benign prostatic hyperplasia reveals increased alterations in apoptotic pathways.

BACKGROUND: Through whole-exome sequencing of 60 formalin-fixed paraffin-embedded Nigerian (NGRn) benign prostatic hyperplasia (BPH) samples, we identified germline and somatic alterations in apoptotic pathways impacting BPH development and progression. Prostate enlargement is a common occurrence in male aging; however, this enlargement can lead to lower urinary tract symptoms that negatively impact quality of life. This impact is disproportionately present in men of African ancestry. BPH pathophysiology is poorly understood and studies examining non-European populations are lacking. METHODS: In this study, NGRn BPH, normal prostate, and prostate cancer (PCa) tumor samples were sequenced and compared to characterize genetic alterations in NGRn BPH. RESULTS: Two hundred and two nonbenign, ClinVar-annotated germline variants were present in NGRn BPH samples. Six genes [BRCA1 (92%), HSD3B1 (85%), TP53 (37%), PMS2 (23%), BARD1 (20%), and BRCA2 (17%)] were altered in at least 10% of samples; however, compared to NGRn normal and tumor, the frequency of alterations in BPH samples showed no significant differences at the gene or variant level. BRCA2_rs11571831 and TP53_rs1042522 germline alterations had a statistically significant co-occurrence interaction in BPH samples. In at least two BPH samples, 173 genes harbored somatic variants known to be clinically actionable. Three genes (COL18A1, KIF16B, and LRP1) showed a statistically significant (p < 0.05) higher frequency in BPH. NGRn BPH also had five gene pairs (PKD1/KIAA0100, PKHD1/PKD1, DNAH9/LRP1B, NWD1/DCHS2, and TCERG1/LMTK2) with statistically significant co-occurring interactions. Two hundred and seventy-nine genes contained novel somatic variants in NGRn BPH. Three genes (CABP1, FKBP1C, and RP11-595B24.2) had a statistically significant (p < 0.05) higher alteration frequency in NGRn BPH and three were significantly higher in NGRn tumor (CACNA1A, DMKN, and CACNA2D2). Pairwise Fisher's exact tests showed 14 gene pairs with statistically significant (p < 0.05) interactions and four interactions approaching significance (p < 0.10). Mutational patterns in NGRn BPH were similar to COSMIC (Catalog of Somatic Mutations in Cancer) signatures associated with aging and dysfunctional DNA damage repair. CONCLUSIONS: NGRn BPH contained significant germline alteration interactions (BRCA2_rs11571831 and TP53_rs1042522) and increased somatic alteration frequencies (LMTK2, LRP1, COL18A1, CABP1, and FKBP1C) that impact apoptosis. Normal prostate development is maintained by balancing apoptotic and proliferative activity. Dysfunction in either mechanism can lead to abnormal prostate growth. This work is the first to examine genomic sequencing in NGRn BPH and provides data that fill known gaps in the understanding BPH and how it impacts men of African ancestry.

Clinical Challenges of Emerging Acquired Autoinflammatory Diseases, Including VEXAS Syndrome.

Vacuoles, E1-ubiquitin-activating enzyme, X-linked, autoinflammatory, somatic (VEXAS) syndrome, caused by an acquired mutation in the ubiquitin-activating enzyme ubiquitin-like modifier activating enzyme 1 (UBA1), was discovered in 2020. Since then, many cases have been reported worldwide. Recently, we performed UBA1 genetic testing in suspected cases of VEXAS throughout Japan and investigated the clinical features of these cases. Most cases were elderly patients in their 70s with clinical features consistent with VEXAS syndrome, such as myelodysplastic syndrome, high-grade fever, skin rash, chondritis, and pulmonary infiltration. However, approximately half of the analyzed patients were negative for the UBA1 variant. As the concept of "acquired autoinflammatory diseases", including VEXAS syndrome, has gained popularity, the number of suspected cases is expected to increase. Currently, there are no established diagnostic or treatment guidelines for these conditions, and they need to be urgently developed. This review summarizes the clinical problems faced by patients with acquired autoinflammatory diseases, including VEXAS.

Presumed Pathogenic Germ Line and Somatic Variants in African American Thyroid Cancer.

Background: African American (AA) thyroid cancer patients have worse prognoses than European Americans (EA), which has been attributed to both health care disparities and possible genetic differences. We investigated the impact of both germ line and somatic variants on clinical outcome in a cohort of AA nonmedullary thyroid cancer (NMTC) patients who had received therapeutic intervention from cancer centers. Methods: Whole-exome sequencing was performed on DNA from available blood/normal tissues (N = 37) and paired tumor samples (N = 32) collected from 37 and 29 AA NMTC patients, respectively. Variants with Combined Annotation Depletion Dependent (CADD) score of ≥20 and VarSome Clinical classification of likely pathogenic or pathogenic were classified as presumed pathogenic germ line or somatic variants (PPGVs/PPSVs). PPGVs/PPSVs in cancer-related genes and PPGVs in cardiovascular risk genes were further investigated, and PPGVs/PPSVs associated with African (AFR) ancestry were identified. Results: Among 17 PPGVs identified in 16 cancer predisposition or known cancer-related genes, only WRN was previously known to associate with NMTC predisposition. Among PPSVs, BRAFV600E was most the prevalent and detected in 12 of the 29 (41%) tumors. Examining PPGVs/PPSVs among three patients who died from NMTC, one patient who died from papillary thyroid carcinoma/anaplastic thyroid carcinoma (PTC/ATC) led us to speculate that the PPGV ERCC4R799W may have increased the risk of PPSV TP53R273H acquisition. Among PPGVs identified in 18 cardiovascular risk genes, PPGVs in SC5NA, GYG1, CBS, CFTR, and SI are known to have causal and pathogenic implications in cardiovascular disease. Conclusion: In this cohort, most AA-NMTC patients exhibit favorable outcomes after therapeutic intervention given at cancer centers, suggesting that health care disparity is the major contributor for worse prognoses among AA-NMTC patients. Nevertheless, the clinical impact of PPGVs that might facilitate the acquisition of TP53 tumor mutations, and/or PPGVs that predispose individuals to adverse cardiovascular events, which could be exacerbated by therapy-induced cardiotoxicity, needs to be further explored. Integrated analysis of PPGV/PPSV profiles among NMTC patients with different stages of disease may help to identify NMTC patients who require close monitoring or proactive intervention.

Integrated exome sequencing and microarray analyses detected genetic defects and underlying pathways of hepatocellular carcinoma.

We performed whole exome sequencing (WES) and microarray analysis to detect somatic variants and copy number alterations (CNAs) for underlying mechanisms in a case series of hepatocellular carcinoma (HCC) with paired DNA samples from tumor and adjacent nontumor tissues. Clinicopathologic findings based on Edmondson-Steiner (E-S) grading, Barcelona-Clinic Liver Cancer (BCLC) stages, recurrence, and survival status and their associations with tumor mutation burden (TMB) and CNA burden (CNAB) were evaluated. WES from 36 cases detected variants in the TP53, AXIN1, CTNNB1, and SMARCA4 genes, amplifications of the AKT3, MYC, and TERT genes, and deletions of the CDH1, TP53, IRF2, RB1, RPL5, and PTEN genes. These genetic defects affecting the p53/cell cycle control, PI3K/Ras, and ß-catenin pathways were observed in approximately 80% of cases. A germline variant in the ALDH2 gene was detected in 52% of the cases. Significantly higher CNAB in patients with poor prognosis by E-S grade III, BCLC stage C, and recurrence than patients with good prognosis by grade III, stage A, grade III and nonrecurrence was noted. Further analysis on a large case series to correlate genomic profiling with clinicopathologic classifications could provide evidence for diagnostic interpretation, prognostic prediction, and target intervention on involved genes and pathways.

Somatic EPAS1 Variants in Pheochromocytoma and Paraganglioma in Patients With Sickle Cell Disease.

CONTEXT: Somatic EPAS1 variants account for 5% to 8% of all pheochromocytoma and paragangliomas (PPGL) but are detected in over 90% of PPGL in patients with congenital cyanotic heart disease, where hypoxemia may select for EPAS1 gain-of-function variants. Sickle cell disease (SCD) is an inherited hemoglobinopathy associated with chronic hypoxia and there are isolated reports of PPGL in patients with SCD, but a genetic link between the conditions has yet to be established. OBJECTIVE: To determine the phenotype and EPAS1 variant status of patients with PPGL and SCD. METHODS: Records of 128 patients with PPGL under follow-up at our center from January 2017 to December 2022 were screened for SCD diagnosis. For identified patients, clinical data and biological specimens were obtained, including tumor, adjacent non-tumor tissue and peripheral blood. Sanger sequencing of exons 9 and 12 of EPAS1, followed by amplicon next-generation sequencing of identified variants was performed on all samples. RESULTS: Four patients with both PPGL and SCD were identified. Median age at PPGL diagnosis was 28 years. Three tumors were abdominal paragangliomas and 1 was a pheochromocytoma. No germline pathogenic variants in PPGL-susceptibility genes were identified in the cohort. Genetic testing of tumor tissue detected unique EPAS1 variants in all 4 patients. Variants were not detected in the germline, and 1 variant was detected in lymph node tissue of a patient with metastatic disease. CONCLUSION: We propose that somatic EPAS1 variants may be acquired through exposure to chronic hypoxia in SCD and drive PPGL development. Future work is needed to further characterize this association.

Germline EGFR c.2527G > A (p.V843I) variant and familial lung cancer.

BACKGROUND: Somatic epidermal growth factor receptor (EGFR) pathogenic variants have been identified and are routinely tested in the molecular diagnosis of non-small cell lung cancer (NSCLC) as they represent a target for EGFR tyrosine kinase inhibitor (TKI) therapy. However, germline variants in EGFR are much less frequently reported. CASE PRESENTATION: Herein, we report the case of a 46-year-old woman diagnosed with lung adenocarcinoma who was found to harbor a rare germline missense variant in exon 21 of EGFR: NM_005228.5(EGFR):c.2527G>A (p.V843I). In the tumor, this variant (Cosmic ID COSV51767379) was accompanied by a secondary, known pathogenic EGFR variant in cis, also occurring in exon 21, c.2573T>G (p.L858R) (Cosmic ID 6224). Her mother was previously diagnosed with poorly differentiated lung carcinoma and her tumor was also found to harbour the p.V843I variant but no other pathogenic variants. Notably, the proband's sister, diagnosed with a lung carcinoma with sarcomatous features at age 44, did not carry this variant or any other somatic or germline EGFR variants. CONCLUSION: This is the second report of familial lung adenocarcinoma associated with the germline p.V843I variant, which remains classified as a variant of uncertain significance. The lack of segregation of this variant in the proband's affected sister illustrates the complexity with evaluating lung cancer predisposition factors. Currently, there is a paucity of data regarding the therapeutic outcomes of patients with tumors expressing this rare germline variant, therefore we propose an algorithm for the identification of at-risk individuals and families as the first step for their personalized management.

Validation of genetic variants from NGS data using deep convolutional neural networks.

Accurate somatic variant calling from next-generation sequencing data is one most important tasks in personalised cancer therapy. The sophistication of the available technologies is ever-increasing, yet, manual candidate refinement is still a necessary step in state-of-the-art processing pipelines. This limits reproducibility and introduces a bottleneck with respect to scalability. We demonstrate that the validation of genetic variants can be improved using a machine learning approach resting on a Convolutional Neural Network, trained using existing human annotation. In contrast to existing approaches, we introduce a way in which contextual data from sequencing tracks can be included into the automated assessment. A rigorous evaluation shows that the resulting model is robust and performs on par with trained researchers following published standard operating procedure.

A study of somatic BRCA variants and their putative effect on protein properties in malignant mesothelioma.

Objectives: The aim of this study is to analyze the prevalence of somatic mutations in BRCA1 and BRCA2 in malignant mesothelioma and their putative impact on protein properties. Methods: Eighteen cases of malignant mesothelioma were retrieved from the archives and for next generation sequencing analysis of BRCA1 and BRCA2 genes. Variants were analyzed using Ensembl VEP17, Polyphen 2.0 software, SIFT software, MutpredV2, and SWISS-MODEL homology-modeling pipeline server. Results: BRCA2 variants were found in significantly higher percentage (22%) of cases (p=0.02). Five missense variants were identified. These were p.A2351P, p.T2250A, p.A895V, pG1771D, and p.R2034C. The SIFT scores of all except one were ≥ 0.03. The Polyphen scores of these four alterations were ≤0.899. In case of p.A2315, the SIFT score was 0.01, while the Polyphen 2 score was 0.921. MutPred2 scores were ≤0.180 for all. Loss of intrinsic disorder was predicted (Pr=0.32, p=0.07) for p.R2034C, while gain of intrinsic disorder was predicted for p.A2351P (Pr=0.36, p=0.01) and p.G1771D (Pr=0.34, p=0.02). Conclusions: BRCA2 somatic variants were identified in 22% cases of malignant mesotheliomas in this study. The variants localize more frequently to the disordered regions of the protein and are predicted to affect the level of disorder.