Halvade somatic: Somatic variant calling with Apache Spark

Background The accurate detection of somatic variants from sequencing data is of key importance for cancer treatment and research. Somatic variant calling requires a high sequencing depth of the tumor sample, especially when the detection of low-frequency variants is also desired. In turn, this leads to large volumes of raw sequencing data to process and hence, large computational requirements. For example, calling the somatic variants according to the GATK best practices guidelines requires days of computing time for a typical whole-genome sequencing sample. Findings We introduce Halvade Somatic, a framework for somatic variant calling from DNA sequencing data that takes advantage of multi-node and/or multi-core compute platforms to reduce runtime. It relies on Apache Spark to provide scalable I/O and to create and manage data streams that are processed on different CPU cores in parallel. Halvade Somatic contains all required steps to process the tumor and matched normal sample according to the GATK best practices recommendations: read alignment (BWA), sorting of reads, preprocessing steps such as marking duplicate reads and base quality score recalibration (GATK), and, finally, calling the somatic variants (Mutect2). Our approach reduces the runtime on a single 36-core node to 19.5 h compared to a runtime of 84.5 h for the original pipeline, a speedup of 4.3 times. Runtime can be further decreased by scaling to multiple nodes, e.g., we observe a runtime of 1.36 h using 16 nodes, an additional speedup of 14.4 times. Halvade Somatic supports variant calling from both whole-genome sequencing and whole-exome sequencing data and also supports Strelka2 as an alternative or complementary variant calling tool. We provide a Docker image to facilitate single-node deployment. Halvade Somatic can be executed on a variety of compute platforms, including Amazon EC2 and Google Cloud. Conclusions To our knowledge, Halvade Somatic is the first somatic variant calling pipeline that leverages Big Data processing platforms and provides reliable, scalable performance. Source code is freely available.

The blended phenotype of a germline RIT1 and a mosaic PIK3CA variant

We report a patient with a germline RIT1 and a mosaic PIK3CA variant. The diagnosis of the RASopathy was confirmed by targeted sequencing following the identification of transient cardiomyopathy in a patient with PIK3CA-related overgrowth spectrum (PROS). Our observation confirms that the PIK3CA gain-of-function (GoF) variant effects dominate those of the RASopathy, and the resulting blended phenotype mostly resembles megalencephaly-capillary malformation syndrome (MCAP PROS). There appears to be interaction between RIT1 and PI3K-AKT because the latter pathway is needed for the growth-promoting activity of the first, at least in adenocarcinomas, but the details of this interaction are not known. If so, the PIK3CA somatic variant may not be just a chance event. It could also be of etiological relevance that Rit activation mediates resistance to cellular stress—that is, promotes cell survival. This anti-apoptotic effect could also make it more likely that a cell that spontaneously acquires a PIK3CA GoF variant will survive and proliferate. We aim to encourage clinicians to investigate atypical findings in individuals with PROS. If further similar cases are reported, this would suggest that the establishment of PROS mosaicism is facilitated by the background of a RASopathy.

Performance Comparison of Computational Prediction Methods for the Function and Pathogenicity of Non-coding Variants

Non-coding variants in the human genome greatly influence some traits and complex diseases by their own regulation and modification effects. Hence, an increasing number of computational methods are developed to predict the effects of variants in the human non-coding sequences. However, it is difficult for users with insufficient knowledge about the performances of computational methods to select appropriate computational methods from dozens of methods. In order to solve this problem, we assessed 12 performance measures of 24 methods on four independent non-coding variant benchmark datasets: (Ⅰ) rare germline variant from ClinVar, (Ⅱ) rare somatic variant from COSMIC, (Ⅲ) common regulatory variant dataset, and (Ⅳ) disease associated common variant dataset. All 24 tested methods performed differently under various conditions, indicating that these methods have varying strengths and weaknesses under different scenarios. Importantly, the performance of existing methods was acceptable in the rare germline variant from ClinVar with area under curves (AUCs) of 0.4481 - 0.8033 and poor in the rare somatic variant from COSMIC (AUCs: 0.4984 - 0.7131), common regulatory variant dataset (AUCs: 0.4837 - 0.6472), and disease associated common variant dataset (AUCs: 0.4766 -0.5188). We also compared the prediction performance among 24 methods for non-coding de novo mutations in autism spectrum disorder and found that the CADD and CDTS methods showed better performance. Summarily, we assessed the performances of 24 computational methods under diverse scenarios, providing preliminary advice for proper tool selection and new method development in interpreting non-coding variants.

Genomic sequencing and functional analyses identify MAP4K3/GLK germline and somatic variants associated with systemic lupus erythematosus

ObjectivesMAP4K3 (GLK) overexpression in T cells induces interleukin (IL)-17A production and autoimmune responses. GLK overexpressing T-cell population is correlated with severity of human systemic lupus erythematosus (SLE); however, it is unclear how GLK is upregulated in patients with SLE.MethodsWe enrolled 181 patients with SLE and 250 individuals without SLE (93 healthy controls and 157 family members of patients with SLE) in two independent cohorts from different hospitals/cities. Genomic DNAs of peripheral blood mononuclear cells were subjected to next-generation sequencing to identify GLK gene variants. The functional consequences of the identified GLK germline or somatic variants were investigated using site-directed mutagenesis and cell transfection, followed by reporter assays, mass spectrometry, immunoblotting, coimmunoprecipitation, and in situ proximity ligation assays.ResultsWe identified 58 patients with SLE from Cohort #1 and #2 with higher frequencies of a somatic variant (chr2:39 477 124 A>G) in GLK 3′-untranslated region (UTR); these patients with SLE showed increased serum anti-double-stranded DNA levels and decreased serum C3/C4 levels. This somatic variant in 3′-UTR enhanced GLK mRNA levels in T cells. In addition, we identified five patients with SLE with GLK (A410T) germline variant in Cohort #1 and #2, as well as two other patients with SLE with GLK (K650R) germline variant in Cohort #1. Another GLK germline variant, A579T, was also detected in one patient with SLE from Cohort #2. Both GLK (A410T) and GLK (K650R) mutants inhibited GLK ubiquitination induced by the novel E3 ligase makorin ring-finger protein 4 (MKRN4), leading to GLK protein stabilisation.ConclusionsMultiple GLK germline and somatic variants cause GLK induction by increasing mRNA or protein stability in patients with SLE.

Next Generation Sequencing of Tumor and Matched Plasma Samples: Identification of Somatic Variants in ctDNA From Ovarian Cancer Patients

Genetic testing to detect somatic alterations is usually performed on formalin-fixed paraffin-embedded tumor samples. However, tumor molecular profiling through ctDNA analysis may be particularly interesting with the emergence of targeted therapies for ovarian cancer (OC), mainly when tumor is not available and biopsy is not viable, also allowing representation of multiple neoplastic subclones. Using a custom panel of 27 genes, next-generation sequencing (NGS) was performed on tumor and matched plasma samples from 96 OC patients, which were combined in two groups (treatment naive and post-treatment). Overall, at least one somatic variant present in the tumor sample was also detected in the matched plasma sample in 35.6% of the patients, a percentage that increased to 69.6% of the treatment naive patients and 83.3% of those with stage IV disease, showing the potential of ctDNA analysis as an alternative to identify somatic variants in these patients, namely those that have predictive value for targeted therapy. In fact, of the two treatment-naive patients with somatic BRCA1 variants identified in tumor samples, in one of them we detected in ctDNA a BRCA1 somatic variant that was present in the tumor with a VAF of 53%, but not in the one that had a VAF of 5.4%. We also showed that ctDNA analysis has a complementary role to molecular unraveling of inter- and intra-tumor heterogeneity, as exemplified by one patient diagnosed with bilateral OC in which different somatic variants from both tumors were detected in ctDNA. Interestingly, as these bilateral tumors shared a rare combination of two of the three variants identified in ctDNA, we could conclude that these morphologically different tumors were clonally related and not synchronous independent neoplasias. Moreover, in the post-treatment group of patients with plasma samples collected after surgery, those with detectable somatic variants had poor prognosis when compared with patients with no detectable somatic variants, highlighting the potential of ctDNA analysis to identify patients at higher risk of recurrence. Concluding, this study demonstrated that somatic variants can be detected in plasma samples of a significant proportion of OC patients, supporting the use of NGS-based ctDNA testing for noninvasive tumor molecular profiling and to stratify patients according to prognosis.

Paired analysis of tumor mutation burden for lung adenocarcinoma and associated idiopathic pulmonary fibrosis

Genetic alterations underlying the development of lung cancer in individuals with idiopathic pulmonary fibrosis (IPF) have remained unclear. To explore whether genetic alterations in IPF tissue contribute to the development of IPF-associated lung cancer, we here evaluated tumor mutation burden (TMB) and somatic variants in 14 paired IPF and tumor samples from patients with IPF-associated lung adenocarcinoma. We also determined TMB for 22 samples of lung adenocarcinoma from patients without IPF. TMB for IPF-associated lung adenocarcinoma was significantly higher than that for matched IPF tissue (median of 2.94 vs. 1.26 mutations/Mb, P = 0.002). Three and 102 somatic variants were detected in IPF and matched lung adenocarcinoma samples, respectively, with only one pair of specimens sharing one somatic variant. TMB for IPF-associated lung adenocarcinoma was similar to that for lung adenocarcinoma samples with driver mutations (median of 2.94 vs. 2.51 mutations/Mb) and lower than that for lung adenocarcinoma samples without known driver mutations (median of 2.94 vs. 5.03 mutations/Mb, P = 0.130) from patients without IPF. Our findings suggest that not only the accumulation of somatic mutations but other factors such as inflammation and oxidative stress might contribute to the development and progression of lung cancer in patients with IPF.

Accuracy of somatic variant detection workflows for whole genome sequencing experiments

Whole genome sequencing (WGS) becomes increasingly important for advancing personalized cancer care, driving not only basic science studies but also entering into clinical applications. Translating raw WGS data into the right clinical decision requires high accuracy of somatic variant detection, therefore novel data analysis methods have to be carefully evaluated. In this work we tested the performance of well-established somatic variant detection workflows: GATK, CPG-WGS, DRAGEN and Strelka2. By utilizing both real data, with well-defined mutations, and synthetic mutations spiked-in into real data, we were able to assess sensitivity and precision of each workflow, for various coverage and tumor purity levels. Individual tools excelled in different evaluation approaches, however the results demonstrated that DRAGEN has the highest overall performance when sensitivity is preferred over precision, and the opposite is true for CGP-WGS. The differences in results obtained using synthetic and real datasets, indicate that benchmarks based only on a single reference set may provide an incomplete picture.

Recurrent KRAS, KIT and SF3B1 mutations in melanoma of the female genital tract

Background Malignant melanoma of the female genital tract is relatively uncommon and accounts for 3–7% of all melanoma localizations. This study aimed to identify driver genes in melanoma of the female genital tract with the purpose of enhancing understanding of disease pathogenesis and identifying potential new therapeutic targets to develop effective therapies. Methods KIT (CD117) and BRAF expression were detected immunohistochemically. Polymerase Chain Reaction (PCR) and Sanger sequencing techniques were performed to identify the mutational status of BRAF, NRAS, KRAS, NF1, KIT, PDGFRA and SF3B1 on 19 melanomas of the female genital tract, paired with 25 cutaneous melanomas, 18 acral melanomas and 11 melanomas of nasal cavity. Results Somatic variant analysis identified KRAS (6/19; 32%) as the most commonly mutated gene, followed by KIT (4/19; 21%), SF3B1 (3/19; 16%) and NRAS (1/19; 5%). None of the cases were found to harbor BRAF, NF1 and PDGFRA mutations in melanomas of the female genital tract. However, none of the cases were found to harbor SF3B1 and KIT mutations in cutaneous melanomas, acral melanomas and melanomas of nasal cavity. Recurrent KIT mutations, as well as mutations in the less frequently mutated genes NRAS and SF3B1, were exclusively detected in vulvovaginal melanomas, but not in tumors arising in the cervix. However, recurrent KRAS mutations were detected in similar frequencies in tumors of the vulva, vagina, and cervix. Additionally, recurrent KRAS and KIT mutations occurred predominantly in polygonal and epithelioid cell types of melanoma in the female genital tract. Immunohistochemistry revealed moderate or strong cytoplasmic CD117 expression in 6 of the 19 cases (31.6%). Conclusions We observed that gynecologic melanoma harbored distinct mutation rates in the KIT, BRAF, SF3B1, KRAS, and NRAS genes. Our findings support the notion that gynecologic melanoma is a distinct entity from non-gynecologic melanoma, and these findings offer insights into future therapeutic options for these patients.

Specific targeting of the KRAS mutational landscape in myeloma as a tool to unveil the elicited anti-tumor activity.

The multiple myeloma (MM) mutational landscape has identified alterations in KRAS as the most recurring somatic variant. Combining DNA and RNA sequencing, we studied 756 patients and observed KRAS as the most frequently mutated gene in patients at diagnosis; in addition, we demonstrated the persistence or de novo occurrence of the KRAS aberration at disease relapse. Small molecule inhibitors targeting KRAS have been developed; however, they are selective for tumors carrying that KRASG12C mutation. Therefore, there is still a need to develop novel therapeutic approaches to target the KRAS mutational events found in other tumor types, including MM. We have used AZD4785, a potent and selective antisense oligonucleotide (ASO) which selectively targets and down-regulates all KRAS isoforms, as a tool to dissect the functional sequelae secondary to KRAS silencing in MM within the context of the bone marrow niche; and demonstrated its ability to significantly silence KRAS, leading to inhibition of MM tumor growth, both in vitro and in vivo, confirming KRAS as a driver and a therapeutic target in MM.