Detecting statistically significant common insertion sites in retroviral insertional mutagenesis screens

Retroviral insertional mutagenesis screens, which identify genes involved in tumor development in mice, have yielded a substantial number of retroviral integration sites, and this number is expected to grow substantially due to the introduction of high-throughput screening techniques. The data of various retroviral insertional mutagenesis screens are compiled in the publicly available Retroviral Tagged Cancer Gene Database (RTCGD). Integrally analyzing these screens for the presence of common insertion sites (CISs, i.e., regions in the genome that have been hit by viral insertions in multiple independent tumors significantly more than expected by chance) requires an approach that corrects for the increased probability of finding false CISs as the amount of available data increases. Moreover, significance estimates of CISs should be established taking into account both the noise, arising from the random nature of the insertion process, as well as the bias, stemming from preferential insertion sites present in the genome and the data retrieval methodology. We introduce a framework, the kernel convolution (KC) framework, to find CISs in a noisy and biased environment using a predefined significance level while controlling the family-wise error (FWE) (the probability of detecting false CISs). Where previous methods use one, two, or three predetermined fixed scales, our method is capable of operating at any biologically relevant scale. This creates the possibility to analyze the CISs in a scale space by varying the width of the CISs, providing new insights in the behavior of CISs across multiple scales. Our method also features the possibility of including models for background bias. Using simulated data, we evaluate the KC framework using three kernel functions, the Gaussian, triangular, and rectangular kernel function. We applied the Gaussian KC to the data from the combined set of screens in the RTCGD and found that 53% of the CISs do not reach the significance threshold in this combined setting. Still, with the FWE under control, application of our method resulted in the discovery of eight novel CISs, which each have a probability less than 5% of being false detections.     

Clonal Expansion Analysis of Transposon Insertions by High-Throughput Sequencing Identifies Candidate Cancer Genes in a PiggyBac Mutagenesis Screen

Somatic transposon mutagenesis in mice is an efficient strategy to investigate the genetic mechanisms of tumorigenesis. The identification of tumor driving transposon insertions traditionally requires the generation of large tumor cohorts to obtain information about common insertion sites. Tumor driving insertions are also characterized by their clonal expansion in tumor tissue, a phenomenon that is facilitated by the slow and evolving transformation process of transposon mutagenesis. We describe here an improved approach for the detection of tumor driving insertions that assesses the clonal expansion of insertions by quantifying the relative proportion of sequence reads obtained in individual tumors. To this end, we have developed a protocol for insertion site sequencing that utilizes acoustic shearing of tumor DNA and Illumina sequencing. We analyzed various solid tumors generated by PiggyBac mutagenesis and for each tumor >10⁶ reads corresponding to >10⁴ insertion sites were obtained. In each tumor, 9 to 25 insertions stood out by their enriched sequence read frequencies when compared to frequencies obtained from tail DNA controls. These enriched insertions are potential clonally expanded tumor driving insertions, and thus identify candidate cancer genes. The candidate cancer genes of our study comprised many established cancer genes, but also novel candidate genes such as Mastermind-like1 (Mamld1) and Diacylglycerolkinase delta (Dgkd). We show that clonal expansion analysis by high-throughput sequencing is a robust approach for the identification of candidate cancer genes in insertional mutagenesis screens on the level of individual tumors.     

Computational identification of insertional mutagenesis targets for cancer gene discovery

Insertional mutagenesis is a potent forward genetic screening technique used to identify candidate cancer genes in mouse model systems. An important, yet unresolved issue in the analysis of these screens, is the identification of the genes affected by the insertions. To address this, we developed Kernel Convolved Rule Based Mapping (KC-RBM). KC-RBM exploits distance, orientation and insertion density across tumors to automatically map integration sites to target genes. We perform the first genome-wide evaluation of the association of insertion occurrences with aberrant gene expression of the predicted targets in both retroviral and transposon data sets. We demonstrate the efficiency of KC-RBM by showing its superior performance over existing approaches in recovering true positives from a list of independently, manually curated cancer genes. The results of this work will significantly enhance the accuracy and speed of cancer gene discovery in forward genetic screens. KC-RBM is available as R-package.     

PCR-based procedures to isolate insertion sites of DNA elements

During the past several years, retroviral insertional mutagenesis has been fruitfully applied to search for genes/pathways involved in tumorigenesis. Techniques used to identify proviral insertion sites are critical for fulfilling these projects. Although a variety of approaches have been described, an improvement over existing methods is required to recover every possible insertion site for cancer gene discovery, so-called saturation analysis. Here, we have described the development of two ligation-mediated PCR variants, SplinkTA-PCR (STA-PCR) and SplinkBlunt-PCR, for efficient isolation of insertion sites in retrovirus-induced leukemia. Our results demonstrated that these two protocols are complementary to each other and that they are better employed in combination for maximal cloning efficiency. These protocols are easy-to-use, reliable and efficient, and are readily applicable to large-scale cloning of insertion sites of provirus and other integrated DNA elements, as well as for detection and cloning of differential insertions unique to drug-resistant cells.     

Characterization of T-DNA insertion sites in Arabidopsis thaliana and the implications for saturation mutagenesis

A key component of a sound functional genomics infrastructure is the availability of a knockout mutant for every gene in the genome. A fruitful approach to systematically knockingout genes in the plant Arabidopsis thaliana has been the use of transferred-DNA (T-DNA) from Agrobacterium tumefaciens as an insertional mutagen. One of the assumptions underlying the use of T-DNA as a mutagen is that the insertion of these DNA elements into the Arabidopsis genome occurs at randomly selected locations. We have directly investigated the distribution of T-DNA insertions sites in populations of transformed Arabidopsis using two different approaches. To begin with, we utilized a polymerase chain reaction (PCR) procedure to systematically catalog the precise locations of all the T-DNA elements inserted within a 65 kb segment of chromosome IV. Of the 47 T-DNA insertions identified, 30% were found within the coding regions of genes. We also documented the insertion of T-DNA elements within the centromeric region of chromosome IV. In addition to these targeted T-DNA screens, we also mapped the genomic locations of 583 randomly chosen T-DNA elements by sequencing the genomic DNA flanking the insertion sites from individual T-DNA-transformed lines. 35% of these randomly chosen T-DNA insertions were located within the coding regions of genes. For comparison, coding sequences account for 44% of the Arabidopsis genome. Our results demonstrate that there is a small bias towards recovering T-DNA insertions within intergenic regions. However, this bias does not limit the utility of T-DNA as an effective insertional mutagen for use in reverse-genetic strategies.     

Retroviral insertions in the VISION database identify molecular pathways in mouse lymphoid leukemia and lymphoma

AKXD recombinant inbred (RI) strains develop a variety of leukemias and lymphomas due to somatically acquired insertions of retroviral DNA into the genome of hematopoetic cells that can mutate cellular proto-oncogenes and tumor suppressor genes. We generated a new set of tumors from nine AKXD RI strains selected for their propensity to develop B-cell tumors, the most common type of human hematopoietic cancers. We employed a PCR technique called viral insertion site amplification (VISA) to rapidly isolate genomic sequence at the site of provirus insertion. Here we describe 550 VISA sequence tags (VSTs) that identify 74 common insertion sites (CISs), of which 21 have not been identified previously. Several suspected proto-oncogenes and tumor suppressor genes lie near CISs, providing supportive evidence for their roles in cancer. Furthermore, numerous previously uncharacterized genes lie near CISs, providing a pool of candidate disease genes for future research. Pathway analysis of candidate genes identified several signaling pathways as common and powerful routes to blood cancer, including Notch, E-protein, NFκB, and Ras signaling. Misregulation of several Notch signaling genes was confirmed by quantitative RT-PCR. Our data suggest that analyses of insertional mutagenesis on a single genetic background are biased toward the identification of cooperating mutations. This tumor collection represents the most comprehensive study of the genetics of B-cell leukemia and lymphoma development in mice. We have deposited the VST sequences, CISs in a genome viewer, histopathology, and molecular tumor typing data in a public web database called VISION (Viral Insertion Sites Identifying Oncogenes), which is located at . Keith C. Weiser and Bin Liu are authors that contributed equally to this work.     

High-throughput linkage analysis of Mutator insertion sites in maize

Insertional mutagenesis is a cornerstone of functional genomics. High-copy transposable element systems such as Mutator ( Mu ) in maize ( Zea mays ) afford the advantage of high forward mutation rates but pose a challenge for identifying the particular element responsible for a given mutation. Several large mutant collections have been generated in Mu -active genetic stocks, but current methods limit the ability to rapidly identify the causal Mu insertions. Here we present a method to rapidly assay Mu insertions that are genetically linked to a mutation of interest. The method combines elements of MuTAIL (thermal asymmetrically interlaced) and amplification of insertion mutagenized sites (AIMS) protocols and is applicable to the analysis of single mutants or to high-throughput analyses of mutant collections. Briefly, genomic DNA is digested with a restriction enzyme and adapters are ligated. Polymerase chain reaction is performed with TAIL cycling parameters, using a fluorescently labeled Mu primer, which results in the preferential amplification and labeling of Mu -containing genomic fragments. Products from a segregating line are analyzed on a capillary sequencer. To recover a fragment of interest, PCR products are cloned and sequenced. Sequences with lengths matching the size of a band that co-segregates with the mutant phenotype represent candidate linked insertion sites, which are then confirmed by PCR. We demonstrate the utility of the method by identifying Mu insertion sites linked to seed-lethal mutations with a preliminary success rate of nearly 50%.     

Estimating abundances of retroviral insertion sites from DNA fragment length data

MOTIVATION: The relative abundance of retroviral insertions in a host genome is important in understanding the persistence and pathogenesis of both natural retroviral infections and retroviral gene therapy vectors. It could be estimated from a sample of cells if only the host genomic sites of retroviral insertions could be directly counted. When host genomic DNA is randomly broken via sonication and then amplified, amplicons of varying lengths are produced. The number of unique lengths of amplicons of an insertion site tends to increase according to its abundance, providing a basis for estimating relative abundance. However, as abundance increases amplicons of the same length arise by chance leading to a non-linear relation between the number of unique lengths and relative abundance. The difficulty in calibrating this relation is compounded by sample-specific variations in the relative frequencies of clones of each length. RESULTS: A likelihood function is proposed for the discrete lengths observed in each of a collection of insertion sites and is maximized with a hybrid expectation-maximization algorithm. Patient data illustrate the method and simulations show that relative abundance can be estimated with little bias, but that variation in highly abundant sites can be large. In replicated patient samples, variation exceeds what the model implies-requiring adjustment as in Efron (2004) or using jackknife standard errors. Consequently, it is advantageous to collect replicate samples to strengthen inferences about relative abundance.     

Regional mutagenesis using Dissociation in maize

We describe genetic screens, molecular methods and web resources newly available to utilize Dissociation (Ds) as an insertional mutagen in maize. Over 1700 Ds elements have been distributed throughout the maize genome to serve as donor elements for local or regional mutagenesis. Two genetic screens are described to identify Ds insertions in genes-of-interest (goi). In scheme I, Ds is used to generate insertion alleles when a recessive reference allele is available. A Ds insertion will enable the cloning of the target gene and can be used to create an allelic series. In scheme II, Ds insertions in a goi are identified using a PCR-based screen to identify the rare insertion alleles among a population of testcross progeny. We detail an inverse PCR protocol to rapidly amplify sequences flanking Ds insertion alleles and describe a high-throughput 96-well plate-based DNA extraction method for the recovery of high-quality genomic DNA from seedling tissues. We also describe several web-based tools for browsing, searching and accessing the genetic materials described. The development of these Ds insertion lines promises to greatly accelerate functional genomics studies in maize.     

Promoter-trapping in Saccharomyces cerevisiae by radiation-assisted fragment insertion

Non-homologous insertion (NHI) of DNA fragments into genomic DNA is a method widely used in insertional mutagenesis screens. In the yeast Saccharomyces cerevisiae, the efficiency of NHI is very low. Here we report that its efficiency can be increased by γ-irradiation of recipient cells at the time of transformation. Radiation-assisted NHI depends on YKU70, but its efficiency is not improved by inactivation of RAD5 or RAD52. In a pilot study, we generated 102 transformant clones expressing a lacZ reporter gene under standard conditions (30°C, rich medium). The site of insertion was determined in a subset of eight clones in which lacZ expression was altered by UV-irradiation. A comparison with published data revealed that three of the eight genes identified in our screen have not been targeted by large-scale transposon-based insertion screens. This suggests that radiation-assisted NHI offers a more homogeneous coverage of the genome than methods relying on transposons or retroviral elements.     

Sequencing methods and datasets to improve functional interpretation of sleeping beauty mutagenesis screens

Background

Animal models of cancer are useful to generate complementary datasets for comparison to human tumor data. Insertional mutagenesis screens, such as those utilizing the Sleeping Beauty (SB) transposon system, provide a model that recapitulates the spontaneous development and progression of human disease. This approach has been widely used to model a variety of cancers in mice. Comprehensive mutation profiles are generated for individual tumors through amplification of transposon insertion sites followed by high-throughput sequencing. Subsequent statistical analyses identify common insertion sites (CISs), which are predicted to be functionally involved in tumorigenesis. Current methods utilized for SB insertion site analysis have some significant limitations. For one, they do not account for transposon footprints – a class of mutation generated following transposon remobilization. Existing methods also discard quantitative sequence data due to uncertainty regarding the extent to which it accurately reflects mutation abundance within a heterogeneous tumor. Additionally, computational analyses generally assume that all potential insertion sites have an equal probability of being detected under non-selective conditions, an assumption without sufficient relevant data. The goal of our study was to address these potential confounding factors in order to enhance functional interpretation of insertion site data from tumors.

Results

We describe here a novel method to detect footprints generated by transposon remobilization, which revealed minimal evidence of positive selection in tumors. We also present extensive characterization data demonstrating an ability to reproducibly assign semi-quantitative information to individual insertion sites within a tumor sample. Finally, we identify apparent biases for detection of inserted transposons in several genomic regions that may lead to the identification of false positive CISs.

Conclusion

The information we provide can be used to refine analyses of data from insertional mutagenesis screens, improving functional interpretation of results and facilitating the identification of genes important in cancer development and progression.

Electronic supplementary material

The online version of this article (doi:10.1186/1471-2164-15-1150) contains supplementary material, which is available to authorized users.     

Large-scale mutagenesis in p19(ARF)- and p53-deficient mice identifies cancer genes and their collaborative networks

p53 and p19(ARF) are tumor suppressors frequently mutated in human tumors. In a high-throughput screen in mice for mutations collaborating with either p53 or p19(ARF) deficiency, we identified 10,806 retroviral insertion sites, implicating over 300 loci in tumorigenesis. This dataset reveals 20 genes that are specifically mutated in either p19(ARF)-deficient, p53-deficient or wild-type mice (including Flt3, mmu-mir-106a-363, Smg6, and Ccnd3), as well as networks of significant collaborative and mutually exclusive interactions between cancer genes. Furthermore, we found candidate tumor suppressor genes, as well as distinct clusters of insertions within genes like Flt3 and Notch1 that induce mutants with different spectra of genetic interactions. Cross species comparative analysis with aCGH data of human cancer cell lines revealed known and candidate oncogenes (Mmp13, Slamf6, and Rreb1) and tumor suppressors (Wwox and Arfrp2). This dataset should prove to be a rich resource for the study of genetic interactions that underlie tumorigenesis.     

Transfer RNA Genes Are Genomic Targets for De Novo Transposition of the Yeast Retrotransposon Ty3

Insertions of the yeast element Ty3 resulting from induced retrotransposition were characterized in order to identify the genomic targets of transposition. The DNA sequences of the junctions between Ty3 and flanking DNA were determined for two insertions of an unmarked element. Each insertion was at position -17 from the 5' end of a tRNA-coding sequence. Ninety-one independent insertions of a marked Ty3 element were studied by Southern blot analysis. Pairs of independent insertions into seven genomic loci accounted for 14 of these insertions. The DNA sequence flanking the insertion site was determined for at least one member of each pair of integrated elements. In each case, insertion was at position -16 or -17 relative to the 5' end of one of seven different tRNA genes. This proportion of genomic loci used twice for Ty3 integration is consistent with that predicted by a Poisson distribution for a number of genomic targets roughly equivalent to the estimated number of yeast tRNA genes. In addition, insertions upstream of the same tRNA gene in one case were at different positions, but in all cases were in the same orientation. Thus, genomic insertions of Ty3 in a particular orientation are apparently specified by the target, while the actual position of the insertion relative to the tRNA-coding sequence can vary slightly.     

Co-occurrence analysis of insertional mutagenesis data reveals cooperating oncogenes

MOTIVATION: Cancers are caused by an accumulation of multiple independent mutations that collectively deregulate cellular pathways, e.g. such as those regulating cell division and cell-death. The publicly available Retroviral Tagged Cancer Gene Database (RTCGD) contains the data of many insertional mutagenesis screens, in which the virally induced mutations result in tumor formation in mice. The insertion loci therefore indicate the location of putative cancer genes. Additionally, the presence of multiple independent insertions within one tumor hints towards a cooperation between the insertionally mutated genes. In this study we focus on the detection of statistically significant co-mutations. RESULTS: We propose a two-dimensional Gaussian Kernel Convolution method (2DGKC), a computational technique that identifies the cooperating mutations in insertional mutagenesis data. We define the Common Co-occurrence of Insertions (CCI), signifying the co-mutations that are statistically significant across all different screens in the RTCGD. Significance estimates are made on multiple scales, and the results visualized in a scale space, thereby providing valuable extra information on the putative cooperation. The multidimensional analysis of the insertion data results in the discovery of 86 statistically significant co-mutations, indicating the presence of cooperating oncogenes that play a role in tumor development. Since oncogenes may cooperate with several members of a parallel pathway, we combined the co-occurrence data with gene family information to find significant cooperations between oncogenes and families of genes. We show, for instance, the interchangeable cooperation of Myc insertions with insertions in the Pim family. AVAILABILITY: A list of the resulting CCIs is available at: http://ict.ewi.tudelft.nl/~jeroen/CCI/CCI_list.txt.