TiGER: A database for tissue-specific gene expression and regulation
© Liu et al; licensee BioMed Central Ltd. 2008
Received: 21 December 2007
Accepted: 09 June 2008
Published: 09 June 2008
Understanding how genes are expressed and regulated in different tissues is a fundamental and challenging question. However, most of currently available biological databases do not focus on tissue-specific gene regulation.
The recent development of computational methods for tissue-specific combinational gene regulation, based on transcription factor binding sites, enables us to perform a large-scale analysis of tissue-specific gene regulation in human tissues. The results are stored in a web database called TiGER (Tissue-specific Gene Expression and Regulation). The database contains three types of data including tissue-specific gene expression profiles, combinatorial gene regulations, and cis-regulatory module (CRM) detections. At present the database contains expression profiles for 19,526 UniGene genes, combinatorial regulations for 7,341 transcription factor pairs and 6,232 putative CRMs for 2,130 RefSeq genes.
We have developed and made publicly available a database, TiGER, which summarizes and provides large scale data sets for tissue-specific gene expression and regulation in a variety of human tissues. This resource is available at .
A detailed understanding of how genes are expressed and regulated in different tissues can help elucidate the molecular mechanisms of tissue development and function. The approximately 25,000 genes in the human genome demonstrate dramatic diversity in terms of expression levels, both temporally and spatially. Despite this diversity, the expression of all genes is controlled by a relatively small number (<2,000) of transcription factors (TFs). These TFs usually work in specific combination to regulate individual genes [2, 3]. A number of databases have been created to facilitate studies of gene expression and regulation. For example, dbEST  is a database of expressed sequence tags (ESTs) from a number of organisms; GNF SymAtlas  and BodyMap  are databases that store human and mouse tissue gene expression profiles; TRANSFAC, TRANSCOMPEL and TRED [7, 8] are databases that store information about transcriptional regulation. Some databases, such as CGED  and PEDB , allow users to access gene expression information derived from either human cancer tissues or one particular tissue (e.g., prostate). However, for a researcher who is interested in tissue-specific gene regulation and would like to examine possible cis-regulatory elements for a gene, a database dedicated to comprehensive information about tissue-specific gene regulation is desirable.
To address this need, we have developed a new database called TiGER (Tissue-specific Gene Expression and Regulation) based on our previous analyses of tissue-specific genes, TFs and cis-regulatory modules (CRMs) for 30 human tissues [3, 11]. TiGER should not be confused with the earlier TIGR databases  on regulation in microbes, plants and humans. TiGER provides simple search engines so as to permit the users to visualize or download information through a standard web browser. More specifically, the TiGER database has the following features:
A large set of data on both tissue-specific genes and tissue-specific transcriptional regulatory elements: The database contains tissue-specific expression profiles for ~20,000 UniGene genes, combinatorial regulation for 7,341 interacting TF pairs, and 6,232 cis-regulatory modules for tissue-specific genes.
Flexible search capability: The database provides three views (gene view, TF view, and tissue view) to allow users to conveniently retrieve information about genes, TFs or tissues of interest. For example, users can simply type a gene ID (e.g., RefSeq) to retrieve the EST profile and CRM detections. Users can also select a tissue name to retrieve a list of genes preferentially expressed in the tissue.
Convenient accessibility: The database provides visualizations of the gene expression profiles, TF interactions and CRM detections. Sortable summary tables, links to raw data and links to external databases are also provided for user reference.
The rest of the paper will describe the database content and illustrate the utility of the database in tissue-specific gene regulation.
Construction and Content
TiGER contains three types of data including tissue-specific gene expression profiles, TF interactions and CRMs. The data are organized as a relational database with a user-friendly interface. The following is a detailed description of the database content.
Tissue-specific gene expression profiles
Tissue specific TF interactions
We have developed a method to identify interacting TFs based on patterns of co-occurrence of pairs of DNA binding sites . This method predicts two TFs interact with each other if their binding sites have over-represented co-occurrence in the promoters of tissue-specific genes and the distances (in unite of base pair) between two sites are significantly different from random expectation (as indicated by a small p-value). Using this method, we predicted 9060 tissue-specific TF interactions, around 300 for each tissue. The predicted interactions include many known TF interactions (e.g., MYOD and MEF2 are known to regulate muscle-specific genes) as well as novel interactions. To evaluate these results, we use known interactions as positive control due to the scarcity of tissue-specific interaction. More than 40% of the known interactions are recovered, with 84-fold enrichment compared to the expected. Figure 1B shows the distribution of -log10(p) values for the 307 eye-specific TF interactions. The most significant is the interaction between FOXJ2 and POU3F2, with a p-value less than 10-39. These two TFs together regulate many eye-specific genes, including RPE65.
Detection of CRMs
CRMs are the central cis-elements that control gene expression . Previously we developed a method to predict CRMs based on TF interactions . This method calculates the interaction strength between two TF binding sites and then derives an empirical "potential energy" for each TF binding site. Using this method, we generated energy profiles for the promoter sequences of tissue-specific genes. An energy level less than -1 indicates the existence of a TF module. We have sensitivity of 12% and enrichment of 10 using known regulatory regions as positive control.
Utility and Discussion
TiGER is constructed for free access and use. The downloadable data formats include standard .txt text files and .png images. The data contents are configured into three views (saved queries): gene view, TF view, and tissue view, to allow users to conveniently retrieve information relevant to genes, TFs or tissues of interest.
There are three major database entities in the gene view: (1) "EST" entity that stores enrichment values in 30 tissues for each gene; (2) "CRM" entity that stores the conservation profile, the density profile, and the energy profile used for CRM detections in the promoter region of each gene; and (3) "GeneCode" entity that stores the mapping between UniGene, RefSeq and gene symbol.
The gene view allows users to retrieve information through a simple search engine by entering a UniGene gene, a RefSeq gene or a gene symbol. The query results include a gene description, a plot of the EST profile, a list of tissues in which the gene is preferentially expressed, a plot of the three profiles used in CRM detection, and download links to the EST and CRM profiles. Links to external databases such as NCBI, UCSC Genome Browser, and GeneCard, are also included for user references.
There is one major database entity called "TF-Partner" in the TF view. This entity stores all factors that interact with a given TF, the tissue in which the interaction occurs and the significance (-log10(p)) of the interaction.
The TF view allows users to retrieve TF interactions by entering a TF name. The query results include a summary table of TF interactions, a link to the raw interaction data, and a pie chart which illustrates the distribution of tissues in which the TF interactions occur.
The tissue view contains three database entities: (1) "TSS-Genes" entity that stores genes preferentially expressed in each of the 30 tissues; (2) "TSS-TFs" entity that stores interactions between TFs in each of the 30 tissues; and (3) "TSS-CRMs" entity that stores CRM modules in the promoter regions of tissue-specific genes. More specifically, the "TSS-Genes" entity contains four attributes including RefSeq gene ID, gene symbol, enrichment values, and descriptions. The "TSS-TFs" entity contains three attributions including the two participating TFs and the significance of interaction. The "TSS-CRMs" entity contains eight attributes including the chromosome ID, RefSeq gene ID, CRM start and end positions, transcription start position and orientation, minimum energy, and a list of TFs that regulate the gene.
The query interfaces (views) are implemented as Java servlets which dynamically query the underlying database entities. TiGER operates under an Apache web server and an Apache Tomcat engine on a SuSe Linux system. The plots of gene expression profiles, TF interactions and CRM detections are pre-generated in Matlab.
We performed a large-scale analysis of gene expression, TF interaction and CRM detection in 30 human tissues. The results are stored in a web-enabled database called TiGER and configured so as to permit users to visualize or download the results through a standard web browser.
There are fundamental issues relating to the computational prediction of human gene regulation. Future research will include both prediction models on gene regulation and analysis tools for interpreting prediction results. As more experimental data accumulates related to the nature of TF-DNA interactions, we plan to further develop our predictions on tissue-specific TF interactions. We also plan to extend our work on CRM detection by relating regulatory elements with temporal (e.g., development) and spatial (e.g., cell types) attributes. As new predictions on tissue-specific gene regulation accumulate, the TiGER database will need to be further expanded and modified. We will update the content of the database on a regular basis. We also plan to develop tools relating TiGER data to other available gene expression and regulation data for integrative analysis.
Availability and requirements
Project name: TiGER
Project home page: 
Operating system(s): Platform independent.
Programming language: none.
License: no restriction.
Any restrictions to use by non-academics: no restriction.
The authors thank the National Institutes of Health for financial supports (EY017589, GM076102, and EY001765), and Mr. and Mrs. Robert and Clarice Smith for a generous gift.
- Yu X, Lin J, Masuda T, Esumi N, Zack DJ, Qian J: Genome-wide prediction and characterization of interactions between transcription factors in Saccharomyces cerevisiae. Nucleic Acids Res 2006, 34(3):917–927. 10.1093/nar/gkj487PubMed CentralView ArticlePubMedGoogle Scholar
- Yu X, Lin J, Zack DJ, Qian J: Computational analysis of tissue-specific combinatorial gene regulation: predicting interaction between transcription factors in human tissues. Nucleic Acids Res 2006, 34(17):4925–4936. 10.1093/nar/gkl595PubMed CentralView ArticlePubMedGoogle Scholar
- Boguski MS, Lowe TM, Tolstoshev CM: dbEST–database for "expressed sequence tags". Nat Genet 1993, 4(4):332–333. 10.1038/ng0893-332View ArticlePubMedGoogle Scholar
- Su AI, Wiltshire T, Batalov S, Lapp H, Ching KA, Block D, Zhang J, Soden R, Hayakawa M, Kreiman G, et al.: A gene atlas of the mouse and human protein-encoding transcriptomes. Proc Natl Acad Sci USA 2004, 101(16):6062–6067. 10.1073/pnas.0400782101PubMed CentralView ArticlePubMedGoogle Scholar
- Hishiki T, Kawamoto S, Morishita S, Okubo K: BodyMap: a human and mouse gene expression database. Nucleic Acids Res 2000, 28(1):136–138. 10.1093/nar/28.1.136PubMed CentralView ArticlePubMedGoogle Scholar
- Matys V, Kel-Margoulis OV, Fricke E, Liebich I, Land S, Barre-Dirrie A, Reuter I, Chekmenev D, Krull M, Hornischer K, et al.: TRANSFAC and its module TRANSCompel: transcriptional gene regulation in eukaryotes. Nucleic Acids Res 2006, (34 Database):D108-D110. 10.1093/nar/gkj143Google Scholar
- Zhao F, Xuan Z, Liu L, Zhang MQ: TRED: a Transcriptional Regulatory Element Database and a platform for in silico gene regulation studies. Nucleic Acids Res 2005, (33 Database):D103-D107.Google Scholar
- Kato K, Yamashita R, Matoba R, Monden M, Noguchi S, Takagi T, Nakai K: Cancer gene expression database (CGED): a database for gene expression profiling with accompanying clinical information of human cancer tissues. Nucleic Acids Res 2005, (33 Database):D533-D536.Google Scholar
- Nelson PS, Pritchard C, Abbott D, Clegg N: The human (PEDB) and mouse (mPEDB) Prostate Expression Databases. Nucleic Acids Res 2002, 30(1):218–220. 10.1093/nar/30.1.218PubMed CentralView ArticlePubMedGoogle Scholar
- Yu X, Lin J, Zack DJ, Qian J: Identification of tissue-specific cis-regulatory modules based on interactions between transcription factors. BMCBioinformatics 2007, 8(1):437. 10.1186/1471-2105-8-437Google Scholar
- TIGR databases[http://www.tigr.org/db.shtml]
- Schuler GD, Boguski MS, Stewart EA, Stein LD, Gyapay G, Rice K, White RE, Rodriguez-Tome P, Aggarwal A, Bajorek E, et al.: A gene map of the human genome. Science 1996, 274(5287):540–546. 10.1126/science.274.5287.540View ArticlePubMedGoogle Scholar
- Istrail S, Davidson EH: Logic functions of the genomic cis-regulatory code. Proc Natl Acad Sci USA 2005, 102(14):4954–4959. 10.1073/pnas.0409624102PubMed CentralView ArticlePubMedGoogle Scholar
- Kent WJ, Sugnet CW, Furey TS, Roskin KM, Pringle TH, Zahler AM, Haussler D: The human genome browser at UCSC. Genome Res 2002, 12(6):996–1006. 10.1101/gr.229102. Article published online before print in May 2002PubMed CentralView ArticlePubMedGoogle Scholar
- Siepel A, Bejerano G, Pedersen JS, Hinrichs AS, Hou M, Rosenbloom K, Clawson H, Spieth J, Hillier LW, Richards S, et al.: Evolutionarily conserved elements in vertebrate, insect, worm, and yeast genomes. Genome Res 2005, 15(8):1034–1050. 10.1101/gr.3715005PubMed CentralView ArticlePubMedGoogle Scholar
This article is published under license to BioMed Central Ltd. This is an Open Access article distributed under the terms of the Creative Commons Attribution License (http://creativecommons.org/licenses/by/2.0), which permits unrestricted use, distribution, and reproduction in any medium, provided the original work is properly cited.