Search

Background: Next-generation sequencing (NGS) allows unbiased, in-depth interrogation of cancer genomes. Many somatic variant callers have been developed yet accurate ascertainment of somatic variants remains a considerable challenge as evidenced by the varying mutation call rates and low concordance among callers. Statistical model-based algorithms that are currently available perform well under ideal scenarios, such as high sequencing depth, homogeneous tumor samples, high somatic variant allele frequency (VAF), but show limited performance with sub-optimal data such as low-pass whole-exome/genome sequencing data. While the goal of any cancer sequencing project is to identify a relevant, and limited, set of somatic variants for further sequence/functional validation, the inherently complex nature of cancer genomes combined with technical issues directly related to sequencing and alignment can affect either the specificity and/or sensitivity of most callers., Results: For these reasons, we developed SNooPer, a versatile machine learning approach that uses Random Forest classification models to accurately call somatic variants in low-depth sequencing data. SNooPer uses a subset of variant positions from the sequencing output for which the class, true variation or sequencing error, is known to train the data-specific model. Here, using a real dataset of 40 childhood acute lymphoblastic leukemia patients, we show how the SNooPer algorithm is not affected by low coverage or low VAFs, and can be used to reduce overall sequencing costs while maintaining high specificity and sensitivity to somatic variant calling. When compared to three benchmarked somatic callers, SNooPer demonstrated the best overall performance., Conclusions: While the goal of any cancer sequencing project is to identify a relevant, and limited, set of somatic variants for further sequence/functional validation, the inherently complex nature of cancer genomes combined with technical issues directly related to sequencing and alignment can affect either the specificity and/or sensitivity of most callers. The flexibility of SNooPer's random forest protects against technical bias and systematic errors, and is appealing in that it does not rely on user-defined parameters. The code and user guide can be downloaded at https://sourceforge.net/projects/snooper/ .

Background: The identification of oncogenic driver mutations has largely relied on the assumption that genes that exhibit more mutations than expected by chance are more likely to play an active role in tumorigenesis. Major cancer sequencing initiatives have therefore focused on recurrent mutations that are more likely to be drivers. However, in specific genetic contexts, low frequency mutations may also be capable of participating in oncogenic processes. Reliable strategies for identifying these rare or even patient-specific (private) mutations are needed in order to elucidate more personalized approaches to cancer diagnosis and treatment., Methods: Here we performed whole-exome sequencing on three cases of childhood pre-B acute lymphoblastic leukemia (cALL), representing three cytogenetically-defined subgroups (high hyperdiploid, t(12;21) translocation, and cytogenetically normal). We applied a data reduction strategy to identify both common and rare/private somatic events with high functional potential. Top-ranked candidate mutations were subsequently validated at high sequencing depth on an independent platform and in vitro expression assays were performed to evaluate the impact of identified mutations on cell growth and survival., Results: We identified 6 putatively damaging non-synonymous somatic mutations among the three cALL patients. Three of these mutations were well-characterized common cALL mutations involved in constitutive activation of the mitogen-activated protein kinase pathway (FLT3 p.D835Y, NRAS p.G13D, BRAF p.G466A). The remaining three patient-specific mutations (ACD p.G223V, DOT1L p.V114F, HCFC1 p.Y103H) were novel mutations previously undescribed in public cancer databases. Cytotoxicity assays demonstrated a protective effect of the ACD p.G223V mutation against apoptosis in leukemia cells. ACD plays a key role in protecting telomeres and recruiting telomerase. Using a telomere restriction fragment assay, we also showed that this novel mutation in ACD leads to increased telomere length in leukemia cells., Conclusion: This study identified ACD as a novel gene involved in cALL and points to a functional role for ACD in enhancing leukemia cell survival. These results highlight the importance of rare/private somatic mutations in understanding cALL etiology, even within well-characterized molecular subgroups.

Background: Acute lymphoblastic leukemia (ALL) is the most common pediatric cancer. While the multi-step model of pediatric leukemogenesis suggests interplay between constitutional and somatic genomes, the role of inherited genetic variability remains largely undescribed. Nonsyndromic familial ALL, although extremely rare, provides the ideal setting to study inherited contributions to ALL. Toward this goal, we sequenced the exomes of a childhood ALL family consisting of mother, father and two non-twinned siblings diagnosed with concordant pre-B hyperdiploid ALL and previously shown to have inherited a rare form of PRDM9, a histone H3 methyltransferase involved in crossing-over at recombination hotspots and Holliday junctions. We postulated that inheritance of additional rare disadvantaging variants in predisposing cancer genes could affect genomic stability and lead to increased risk of hyperdiploid ALL within this family., Methods: Whole exomes were captured using Agilent's SureSelect kit and sequenced on the Life Technologies SOLiD System. We applied a data reduction strategy to identify candidate variants shared by both affected siblings. Under a recessive disease model, we focused on rare non-synonymous or frame-shift variants in leukemia predisposing pathways., Results: Though the family was nonsyndromic, we identified a combination of rare variants in Fanconi anemia (FA) genes FANCP/SLX4 (compound heterozygote - rs137976282/rs79842542) and FANCA (rs61753269) and a rare homozygous variant in the Holliday junction resolvase GEN1 (rs16981869). These variants, predicted to affect protein function, were previously identified in familial breast cancer cases. Based on our in-house database of 369 childhood ALL exomes, the sibs were the only patients to carry this particularly rare combination and only a single hyperdiploid patient was heterozygote at both FANCP/SLX4 positions, while no FANCA variant allele carriers were identified. FANCA is the most commonly mutated gene in FA and is essential for resolving DNA interstrand cross-links during replication. FANCP/SLX4 and GEN1 are involved in the cleavage of Holliday junctions and their mutated forms, in combination with the rare allele of PRDM9, could alter Holliday junction resolution leading to nondisjunction of chromosomes and segregation defects., Conclusion: Taken together, these results suggest that concomitant inheritance of rare variants in FANCA, FANCP/SLX4 and GEN1 on the specific genetic background of this familial case, could lead to increased genomic instability, hematopoietic dysfunction, and higher risk of childhood leukemia.

The joint sequencing of related genomes has become an important means to discover rare variants. Normal-tumor genome pairs are routinely sequenced together to find somatic mutations and their associations with different cancers. Parental and sibling genomes reveal de novo germline mutations and inheritance patterns related to Mendelian diseases.Acute lymphoblastic leukemia (ALL) is the most common paediatric cancer and the leading cause of cancer-related death among children. With the aim of uncovering the full spectrum of germline and somatic genetic alterations in childhood ALL genomes, we conducted whole-exome re-sequencing on a unique cohort of over 120 exomes of childhood ALL quartets, each comprising a patient's tumor and matched-normal material, and DNA from both parents. We developed a general probabilistic model for such quartet sequencing reads mapped to the reference human genome. The model is used to infer joint genotypes at homologous loci across a normal-tumor genome pair and two parental genomes.We describe the algorithms and data structures for genotype inference, model parameter training. We implemented the methods in an open-source software package (QUADGT) that uses the standard file formats of the 1000 Genomes Project. Our method's utility is illustrated on quartets from the ALL cohort.