The FANTOM5 project investigates transcription initiation activities in more than 1,000 human and mouse primary cells, cell lines and tissues using CAGE. Based on manual curation of sample information and development of an ontology for sample classification, we assemble the resulting data into a centralized data resource (http://fantom.gsc.riken.jp/5/). This resource contains web-based tools and data-access points for the research community to search and extract data related to samples, genes, promoter activities, transcription factors and enhancers across the FANTOM5 atlas.

Mobilization of retrotransposons, genetic elements able to move around in the genome where they can become incorporated and start to amplify themselves, is normally suppressed in somatic cells. However, recent reports indicate that L1 retrotransposons can be mobilized in the human brain; this has important consequences for intercellular variation. Using a high-throughput approach, Baillie et al. identify numerous germ-line mutations and putative somatic insertions in the human hippocampus and caudate nucleus, including those of Alu elements. The implication is that retrotransposition-driven somatic mosaicism may reshape the genetic circuitry that underpins normal and abnormal neurobiological processes. Retrotransposons are mobile genetic elements that use a germline ‘copy-and-paste’ mechanism to spread throughout metazoan genomes1. At least 50 per cent of the human genome is derived from retrotransposons, with three active families (L1, Alu and SVA) associated with insertional mutagenesis and disease2,3. Epigenetic and post-transcriptional suppression block retrotransposition in somatic cells4,5, excluding early embryo development and some malignancies6,7. Recent reports of L1 expression8,9 and copy number variation10,11 in the human brain suggest that L1 mobilization may also occur during later development. However, the corresponding integration sites have not been mapped. Here we apply a high-throughput method to identify numerous L1, Alu and SVA germline mutations, as well as 7,743 putative somatic L1 insertions, in the hippocampus and caudate nucleus of three individuals. Surprisingly, we also found 13,692 somatic Alu insertions and 1,350 SVA insertions. Our results demonstrate that retrotransposons mobilize to protein-coding genes differentially expressed and active in the brain. Thus, somatic genome mosaicism driven by retrotransposition may reshape the genetic circuitry that underpins normal and abnormal neurobiological processes.

The specialisation of mammalian cells in time and space requires genes associated with specific pathways and functions to be co-ordinately expressed. Here we have combined a large number of publically available microarray datasets derived from human primary cells and analysed large correlation graphs of these data.

Network analysis transcends conventional pairwise approaches to data analysis as the context of components in a network graph can be taken into account. Such approaches are increasingly being applied to genomics data, where functional linkages are used to connect genes or proteins. However, while microarray gene expression datasets are now abundant and of high quality, few approaches have been developed for analysis of such data in a network context. We present a novel approach for 3-D visualisation and analysis of transcriptional networks generated from microarray data. These networks consist of nodes representing transcripts connected by virtue of their expression profile similarity across multiple conditions. Analysing genome-wide gene transcription across 61 mouse tissues, we describe the unusual topography of the large and highly structured networks produced, and demonstrate how they can be used to visualise, cluster, and mine large datasets. This approach is fast, intuitive, and versatile, and allows the identification of biological relationships that may be missed by conventional analysis techniques. This work has been implemented in a freely available open-source application named BioLayout Express^3D. This paper describes a novel approach for analysis of gene expression data. In this approach, normalized gene expression data is transformed into a graph where nodes in the graph represent transcripts connected to each other by virtue of their coexpression across multiple tissues or samples. The graph paradigm has many advantages for such analyses. Graph clustering of the derived network performs extremely well in comparison to traditional pairwise schemes. We show that this approach is robust and able to accommodate large datasets such as the Genomics Institute of the Novartis Research Foundation mouse tissue atlas. The entire approach and algorithms are combined into a single open-source JAVA application that allows users to perform this analysis and further mining on their own data and to visualize the results interactively in 3-D. The approach is not limited to gene expression data but would also be useful for other complex biological datasets. We use the method to investigate the relationship between the phylogenetic age of transcripts and their tissue specificity.

Background: The domestic pig is known as an excellent model for human immunology and the two species share many pathogens. Susceptibility to infectious disease is one of the major constraints on swine performance, yet the structure and function of genes comprising the pig immunome are not well-characterized. The completion of the pig genome provides the opportunity to annotate the pig immunome, and compare and contrast pig and human immune systems. Results: The Immune Response Annotation Group (IRAG) used computational curation and manual annotation of the swine genome assembly 10.2 (Sscrofa10.2) to refine the currently available automated annotation of 1,369 immunity-related genes through sequence-based comparison to genes in other species. Within these genes, we annotated 3,472 transcripts. Annotation provided evidence for gene expansions in several immune response families, and identified artiodactyl-specific expansions in the cathelicidin and type 1 Interferon families. We found gene duplications for 18 genes, including 13 immune response genes and five non-immune response genes discovered in the annotation process. Manual annotation provided evidence for many new alternative splice variants and 8 gene duplications. Over 1,100 transcripts without porcine sequence evidence were detected using cross-species annotation. We used a functional approach to discover and accurately annotate porcine immune response genes. A co-expression clustering analysis of transcriptomic data from selected experimental infections or immune stimulations of blood, macrophages or lymph nodes identified a large cluster of genes that exhibited a correlated positive response upon infection across multiple pathogens or immune stimuli. Interestingly, this gene cluster (cluster 4) is enriched for known general human immune response genes, yet contains many un-annotated porcine genes. A phylogenetic analysis of the encoded proteins of cluster 4 genes showed that 15% exhibited an accelerated evolution as compared to 4.1% across the entire genome. Conclusions: This extensive annotation dramatically extends the genome-based knowledge of the molecular genetics and structure of a major portion of the porcine immunome. Our complementary functional approach using co-expression during immune response has provided new putative immune response annotation for over 500 porcine genes. Our phylogenetic analysis of this core immunome cluster confirms rapid evolutionary change in this set of genes, and that, as in other species, such genes are important components of the pig’s adaptation to pathogen challenge over evolutionary time. These comprehensive and integrated analyses increase the value of the porcine genome sequence and provide important tools for global analyses and data-mining of the porcine immune response.

The immune composition of the tumor microenvironment regulates processes including angiogenesis, metastasis, and the response to drugs or immunotherapy. To facilitate the characterization of the immune component of tumors from transcriptomics data, a number of immune cell transcriptome signatures have been reported that are made up of lists of marker genes indicative of the presence a given immune cell population. The majority of these gene signatures have been defined through analysis of isolated blood cells. However, blood cells do not reflect the differentiation or activation state of similar cells within tissues, including tumors, and consequently markers derived from blood cells do not necessarily transfer well to tissues. To address this issue, we generated a set of immune gene signatures derived directly from tissue transcriptomics data using a network-based deconvolution approach. We define markers for seven immune cell types, collectively named ImSig, and demonstrate how these markers can be used for the quantitative estimation of the immune cell content of tumor and nontumor tissue samples. The utility of ImSig is demonstrated through the stratification of melanoma patients into subgroups of prognostic significance and the identification of immune cells with the use of single-cell RNA-sequencing data derived from tumors. Use of ImSig is facilitated by an R package (imsig). Cancer Immunol Res; 6(11); 1388–400. ©2018 AACR.