Prediction of tumor-specific splicing from somatic mutations as a source of neoantigen candidates.

Motivation: Neoantigens are promising targets for cancer immunotherapies and might arise from alternative splicing. However, detecting tumor-specific splicing is challenging because many non-canonical splice junctions identified in tumors also appear in healthy tissues. To increase tumor-specificity, we focused on splicing caused by somatic mutations as a source for neoantigen candidates in individual patients. Results: We developed the tool splice2neo with multiple functionalities to integrate predicted splice effects from somatic mutations with splice junctions detected in tumor RNA-seq and to annotate the resulting transcript and peptide sequences. Additionally, we provide the tool EasyQuant for targeted RNA-seq read mapping to candidate splice junctions. Using a stringent detection rule, we predicted 1.7 splice junctions per patient as splice targets with a false discovery rate below 5% in a melanoma cohort. We confirmed tumor-specificity using independent, healthy tissue samples. Furthermore, using tumor-derived RNA, we confirmed individual exon-skipping events experimentally. Most target splice junctions encoded neoepitope candidates with predicted major histocompatibility complex (MHC)-I or MHC-II binding. Compared to neoepitope candidates from non-synonymous point mutations, the splicing-derived MHC-I neoepitope candidates had lower self-similarity to corresponding wild-type peptides. In conclusion, we demonstrate that identifying mutation-derived, tumor-specific splice junctions can lead to additional neoantigen candidates to expand the target repertoire for cancer immunotherapies. Availability and implementation: The R package splice2neo and the python package EasyQuant are available at https://github.com/TRON-Bioinformatics/splice2neo and https://github.com/TRON-Bioinformatics/easyquant, respectively.

Accurate detection of tumor-specific gene fusions reveals strongly immunogenic personal neo-antigens.

Cancer-associated gene fusions are a potential source for highly immunogenic neoantigens, but the lack of computational tools for accurate, sensitive identification of personal gene fusions has limited their targeting in personalized cancer immunotherapy. Here we present EasyFuse, a machine learning computational pipeline for detecting cancer-specific gene fusions in transcriptome data obtained from human cancer samples. EasyFuse predicts personal gene fusions with high precision and sensitivity, outperforming previously described tools. By testing immunogenicity with autologous blood lymphocytes from patients with cancer, we detected pre-established CD4+ and CD8+ T cell responses for 10 of 21 (48%) and for 1 of 30 (3%) identified gene fusions, respectively. The high frequency of T cell responses detected in patients with cancer supports the relevance of individual gene fusions as neoantigens that might be targeted in personalized immunotherapies, especially for tumors with low mutation burden.

The developmental transcriptome for Lytechinus variegatus exhibits temporally punctuated gene expression changes.

Embryonic development is arguably the most complex process an organism undergoes during its lifetime, and understanding this complexity is best approached with a systems-level perspective. The sea urchin has become a highly valuable model organism for understanding developmental specification, morphogenesis, and evolution. As a non-chordate deuterostome, the sea urchin occupies an important evolutionary niche between protostomes and vertebrates. Lytechinus variegatus (Lv) is an Atlantic species that has been well studied, and which has provided important insights into signal transduction, patterning, and morphogenetic changes during embryonic and larval development. The Pacific species, Strongylocentrotus purpuratus (Sp), is another well-studied sea urchin, particularly for gene regulatory networks (GRNs) and cis-regulatory analyses. A well-annotated genome and transcriptome for Sp are available, but similar resources have not been developed for Lv. Here, we provide an analysis of the Lv transcriptome at 11 timepoints during embryonic and larval development. Temporal analysis suggests that the gene regulatory networks that underlie specification are well-conserved among sea urchin species. We show that the major transitions in variation of embryonic transcription divide the developmental time series into four distinct, temporally sequential phases. Our work shows that sea urchin development occurs via sequential intervals of relatively stable gene expression states that are punctuated by abrupt transitions.

The distributions of protein coding genes within chromatin domains in relation to human disease.

BACKGROUND: Our understanding of the nuclear chromatin structure has increased hugely during the last years mainly as a consequence of the advances in chromatin conformation capture methods like Hi-C. The unprecedented resolution of genome-wide interaction maps shows functional consequences that extend the initial thought of an efficient DNA packaging mechanism: gene regulation, DNA repair, chromosomal translocations and evolutionary rearrangements seem to be only the peak of the iceberg. One key concept emerging from this research is the topologically associating domains (TADs) whose functional role in gene regulation and their association with disease is not fully untangled. RESULTS: We report that the lower the number of protein coding genes inside TADs, the higher the tendency of those genes to be associated with disease (p-value = 4 × [Formula: see text]). Moreover, housekeeping genes are less associated with disease than other genes. Accordingly, they are depleted in TADs containing less than three protein coding genes (p-value = 3.9 × [Formula: see text]). We observed that TADs with higher ratios of enhancers versus genes contained higher numbers of disease-associated genes. We interpret these results as an indication that sharing enhancers among genes reduces their involvement in disease. Larger TADs would have more chances to accommodate many genes and select for enhancer sharing along evolution. CONCLUSIONS: Genes associated with human disease do not distribute randomly over the TADs. Our observations suggest general rules that confer functional stability to TADs, adding more evidence to the role of TADs as regulatory units.

7C: Computational Chromosome Conformation Capture by Correlation of ChIP-seq at CTCF motifs.

BACKGROUND: Knowledge of the three-dimensional structure of the genome is necessary to understand how gene expression is regulated. Recent experimental techniques such as Hi-C or ChIA-PET measure long-range chromatin interactions genome-wide but are experimentally elaborate, have limited resolution and such data is only available for a limited number of cell types and tissues. RESULTS: While ChIP-seq was not designed to detect chromatin interactions, the formaldehyde treatment in the ChIP-seq protocol cross-links proteins with each other and with DNA. Consequently, also regions that are not directly bound by the targeted TF but interact with the binding site via chromatin looping are co-immunoprecipitated and sequenced. This produces minor ChIP-seq signals at loop anchor regions close to the directly bound site. We use the position and shape of ChIP-seq signals around CTCF motif pairs to predict whether they interact or not. We implemented this approach in a prediction method, termed Computational Chromosome Conformation Capture by Correlation of ChIP-seq at CTCF motifs (7C). We applied 7C to all CTCF motif pairs within 1 Mb in the human genome and validated predicted interactions with high-resolution Hi-C and ChIA-PET. A single ChIP-seq experiment from known architectural proteins (CTCF, Rad21, Znf143) but also from other TFs (like TRIM22 or RUNX3) predicts loops accurately. Importantly, 7C predicts loops in cell types and for TF ChIP-seq datasets not used in training. CONCLUSION: 7C predicts chromatin loops which can help to associate TF binding sites to regulated genes. Furthermore, profiling of hundreds of ChIP-seq datasets results in novel candidate factors functionally involved in chromatin looping. Our method is available as an R/Bioconductor package: http://bioconductor.org/packages/sevenC .

Computational Processing and Quality Control of Hi-C, Capture Hi-C and Capture-C Data.

Hi-C, capture Hi-C (CHC) and Capture-C have contributed greatly to our present understanding of the three-dimensional organization of genomes in the context of transcriptional regulation by characterizing the roles of topological associated domains, enhancer promoter loops and other three-dimensional genomic interactions. The analysis is based on counts of chimeric read pairs that map to interacting regions of the genome. However, the processing and quality control presents a number of unique challenges. We review here the experimental and computational foundations and explain how the characteristics of restriction digests, sonication fragments and read pairs can be exploited to distinguish technical artefacts from valid read pairs originating from true chromatin interactions.

Evolutionary stability of topologically associating domains is associated with conserved gene regulation.

BACKGROUND: The human genome is highly organized in the three-dimensional nucleus. Chromosomes fold locally into topologically associating domains (TADs) defined by increased intra-domain chromatin contacts. TADs contribute to gene regulation by restricting chromatin interactions of regulatory sequences, such as enhancers, with their target genes. Disruption of TADs can result in altered gene expression and is associated to genetic diseases and cancers. However, it is not clear to which extent TAD regions are conserved in evolution and whether disruption of TADs by evolutionary rearrangements can alter gene expression. RESULTS: Here, we hypothesize that TADs represent essential functional units of genomes, which are stable against rearrangements during evolution. We investigate this using whole-genome alignments to identify evolutionary rearrangement breakpoints of different vertebrate species. Rearrangement breakpoints are strongly enriched at TAD boundaries and depleted within TADs across species. Furthermore, using gene expression data across many tissues in mouse and human, we show that genes within TADs have more conserved expression patterns. Disruption of TADs by evolutionary rearrangements is associated with changes in gene expression profiles, consistent with a functional role of TADs in gene expression regulation. CONCLUSIONS: Together, these results indicate that TADs are conserved building blocks of genomes with regulatory functions that are often reshuffled as a whole instead of being disrupted by rearrangements.

Computational Prediction of Position Effects of Apparently Balanced Human Chromosomal Rearrangements.

Interpretation of variants of uncertain significance, especially chromosomal rearrangements in non-coding regions of the human genome, remains one of the biggest challenges in modern molecular diagnosis. To improve our understanding and interpretation of such variants, we used high-resolution three-dimensional chromosomal structural data and transcriptional regulatory information to predict position effects and their association with pathogenic phenotypes in 17 subjects with apparently balanced chromosomal abnormalities. We found that the rearrangements predict disruption of long-range chromatin interactions between several enhancers and genes whose annotated clinical features are strongly associated with the subjects' phenotypes. We confirm gene-expression changes for a couple of candidate genes to exemplify the utility of our analysis of position effect. These results highlight the important interplay between chromosomal structure and disease and demonstrate the need to utilize chromatin conformational data for the prediction of position effects in the clinical interpretation of non-coding chromosomal rearrangements.

Co-regulation of paralog genes in the three-dimensional chromatin architecture.

Paralog genes arise from gene duplication events during evolution, which often lead to similar proteins that cooperate in common pathways and in protein complexes. Consequently, paralogs show correlation in gene expression whereby the mechanisms of co-regulation remain unclear. In eukaryotes, genes are regulated in part by distal enhancer elements through looping interactions with gene promoters. These looping interactions can be measured by genome-wide chromatin conformation capture (Hi-C) experiments, which revealed self-interacting regions called topologically associating domains (TADs). We hypothesize that paralogs share common regulatory mechanisms to enable coordinated expression according to TADs. To test this hypothesis, we integrated paralogy annotations with human gene expression data in diverse tissues, genome-wide enhancer-promoter associations and Hi-C experiments in human, mouse and dog genomes. We show that paralog gene pairs are enriched for co-localization in the same TAD, share more often common enhancer elements than expected and have increased contact frequencies over large genomic distances. Combined, our results indicate that paralogs share common regulatory mechanisms and cluster not only in the linear genome but also in the three-dimensional chromatin architecture. This enables concerted expression of paralogs over diverse cell-types and indicate evolutionary constraints in functional genome organization.

Q-nexus: a comprehensive and efficient analysis pipeline designed for ChIP-nexus.

BACKGROUND: ChIP-nexus, an extension of the ChIP-exo protocol, can be used to map the borders of protein-bound DNA sequences at nucleotide resolution, requires less input DNA and enables selective PCR duplicate removal using random barcodes. However, the use of random barcodes requires additional preprocessing of the mapping data, which complicates the computational analysis. To date, only a very limited number of software packages are available for the analysis of ChIP-exo data, which have not yet been systematically tested and compared on ChIP-nexus data. RESULTS: Here, we present a comprehensive software package for ChIP-nexus data that exploits the random barcodes for selective removal of PCR duplicates and for quality control. Furthermore, we developed bespoke methods to estimate the width of the protected region resulting from protein-DNA binding and to infer binding positions from ChIP-nexus data. Finally, we applied our peak calling method as well as the two other methods MACE and MACS2 to the available ChIP-nexus data. CONCLUSIONS: The Q-nexus software is efficient and easy to use. Novel statistics about duplication rates in consideration of random barcodes are calculated. Our method for the estimation of the width of the protected region yields unbiased signatures that are highly reproducible for biological replicates and at the same time very specific for the respective factors analyzed. As judged by the irreproducible discovery rate (IDR), our peak calling algorithm shows a substantially better reproducibility. An implementation of Q-nexus is available at http://charite.github.io/Q/ .

RNA-Seq identifies SPGs as a ventral skeletal patterning cue in sea urchins.

The sea urchin larval skeleton offers a simple model for formation of developmental patterns. The calcium carbonate skeleton is secreted by primary mesenchyme cells (PMCs) in response to largely unknown patterning cues expressed by the ectoderm. To discover novel ectodermal cues, we performed an unbiased RNA-Seq-based screen and functionally tested candidates; we thereby identified several novel skeletal patterning cues. Among these, we show that SLC26a2/7 is a ventrally expressed sulfate transporter that promotes a ventral accumulation of sulfated proteoglycans, which is required for ventral PMC positioning and skeletal patterning. We show that the effects of SLC perturbation are mimicked by manipulation of either external sulfate levels or proteoglycan sulfation. These results identify novel skeletal patterning genes and demonstrate that ventral proteoglycan sulfation serves as a positional cue for sea urchin skeletal patterning.

ChIP-exo signal associated with DNA-binding motifs provides insight into the genomic binding of the glucocorticoid receptor and cooperating transcription factors.

The classical DNA recognition sequence of the glucocorticoid receptor (GR) appears to be present at only a fraction of bound genomic regions. To identify sequences responsible for recruitment of this transcription factor (TF) to individual loci, we turned to the high-resolution ChIP-exo approach. We exploited this signal by determining footprint profiles of TF binding at single-base-pair resolution using ExoProfiler, a computational pipeline based on DNA binding motifs. When applied to our GR and the few available public ChIP-exo data sets, we find that ChIP-exo footprints are protein- and recognition sequence-specific signatures of genomic TF association. Furthermore, we show that ChIP-exo captures information about TFs other than the one directly targeted by the antibody in the ChIP procedure. Consequently, the shape of the ChIP-exo footprint can be used to discriminate between direct and indirect (tethering to other DNA-bound proteins) DNA association of GR. Together, our findings indicate that the absence of classical recognition sequences can be explained by direct GR binding to a broader spectrum of sequences than previously known, either as a homodimer or as a heterodimer binding together with a member of the ETS or TEAD families of TFs, or alternatively by indirect recruitment via FOX or STAT proteins. ChIP-exo footprints also bring structural insights and locate DNA:protein cross-link points that are compatible with crystal structures of the studied TFs. Overall, our generically applicable footprint-based approach uncovers new structural and functional insights into the diverse ways of genomic cooperation and association of TFs.

Deletions of chromosomal regulatory boundaries are associated with congenital disease.

BACKGROUND: Recent data from genome-wide chromosome conformation capture analysis indicate that the human genome is divided into conserved megabase-sized self-interacting regions called topological domains. These topological domains form the regulatory backbone of the genome and are separated by regulatory boundary elements or barriers. Copy-number variations can potentially alter the topological domain architecture by deleting or duplicating the barriers and thereby allowing enhancers from neighboring domains to ectopically activate genes causing misexpression and disease, a mutational mechanism that has recently been termed enhancer adoption. RESULTS: We use the Human Phenotype Ontology database to relate the phenotypes of 922 deletion cases recorded in the DECIPHER database to monogenic diseases associated with genes in or adjacent to the deletions. We identify combinations of tissue-specific enhancers and genes adjacent to the deletion and associated with phenotypes in the corresponding tissue, whereby the phenotype matched that observed in the deletion. We compare this computationally with a gene-dosage pathomechanism that attempts to explain the deletion phenotype based on haploinsufficiency of genes located within the deletions. Up to 11.8% of the deletions could be best explained by enhancer adoption or a combination of enhancer adoption and gene-dosage effects. CONCLUSIONS: Our results suggest that enhancer adoption caused by deletions of regulatory boundaries may contribute to a substantial minority of copy-number variation phenotypes and should thus be taken into account in their medical interpretation.

Clinical interpretation of CNVs with cross-species phenotype data.

BACKGROUND: Clinical evaluation of CNVs identified via techniques such as array comparative genome hybridisation (aCGH) involves the inspection of lists of known and unknown duplications and deletions with the goal of distinguishing pathogenic from benign CNVs. A key step in this process is the comparison of the individual's phenotypic abnormalities with those associated with Mendelian disorders of the genes affected by the CNV. However, because often there is not much known about these human genes, an additional source of data that could be used is model organism phenotype data. Currently, almost 6000 genes in mouse and zebrafish are, when knocked out, associated with a phenotype in the model organism, but no disease is known to be caused by mutations in the human ortholog. Yet, searching model organism databases and comparing model organism phenotypes with patient phenotypes for identifying novel disease genes and medical evaluation of CNVs is hindered by the difficulty in integrating phenotype information across species and the lack of appropriate software tools. METHODS: Here, we present an integrated ranking scheme based on phenotypic matching, degree of overlap with known benign or pathogenic CNVs and the haploinsufficiency score for the prioritisation of CNVs responsible for a patient's clinical findings. RESULTS: We show that this scheme leads to significant improvements compared with rankings that do not exploit phenotypic information. We provide a software tool called PhenogramViz, which supports phenotype-driven interpretation of aCGH findings based on multiple data sources, including the integrated cross-species phenotype ontology Uberpheno, in order to visualise gene-to-phenotype relations. CONCLUSIONS: Integrating and visualising cross-species phenotype information on the affected genes may help in routine diagnostics of CNVs.