Opinion
References 1 Burns, T.F. and El-Deiry, W.S. (1999) The p53 pathway and apoptosis. J. Cell. Physiol. 181, 231–239 2 Boulaire, J. et al. (2000) The functions of the cdk–cyclin kinase inhibitor p21WAF1. Pathol. Biol. 48, 190–202 3 Offer, H. et al. (1999) Direct involvement of p53 in the base excision repair pathway of the DNA repair machinery. FEBS Lett. 450, 197–204 4 Armando, A-A. et al. (1999) p53 is a rate-limiting factor in the repair of higher-order DNA structure. Biochim. Biophys. Acta 1446, 181–192 5 Smith, M.L. et al. (2000) p53-mediated DNA repair responses to UV radiation: studies of mouse cells lacking p53, p21, and/or gadd45 genes. Mol. Cell. Biol. 20, 3705–3714 6 Cheah, K.S.E. and Osborne, D.J. (1978) DNA lesions occur with loss of viability in embryos of aging rye seed. Nature 272, 593–599 7 Dell’Aquila, A. and Tritto, V. (1990) Ageing and osmotic priming in wheat seeds: effects upon certain components of seed quality. Ann. Bot. 65, 21–26
TRENDS in Plant Science Vol.6 No.6 June 2001
8 Osborne, D.J. (1983) Biochemical control systems operating in the early hours of germination. Can. J. Bot. 61, 3568–3577 9 Kastan, M.B. et al. (1991) Participation of p53 protein in the cellular response to DNA damage. Cancer Res. 51, 6304–6311 10 Wang, B.S.P. and Berjak, P. (2000) Beneficial effects of moist chilling on the seeds of black spruce (Picea mariana (Mill.) B.S.P.). Ann. Bot. 86, 29–36 11 Roberts, E.H. (1988) Seed aging: the genome and its expression. In Senescence and Aging in Plants (Nooden, L.D. and Leopold, A.C., eds), pp. 465–498, Academic Press 12 Sivritepe, H.O. and Dourado, A.M. (1995) The effect of priming treatments on the viability and accumulation of chromosomal damage in aged pea seeds. Ann. Bot. 75, 165–171 13 Georgieva, E.I. et al. (1994) Maize embryo germination II. Proteins related to nuclear protooncogene- and tumor suppressor gene-products. Planta 192, 125–129 14 Cruz-García, F. et al. (1998) Effect of stimulating
Database-assisted promoter analysis Reinhard Hehl and Edgar Wingender The analysis of regulatory sequences is greatly facilitated by database-assisted bioinformatic approaches. The TRANSFAC database contains information on transcription factors and their origins, functional properties and sequencespecific binding activities. Software tools enable us to screen the database with a given DNA sequence for interacting transcription factors. If a regulatory function is already attributed to this sequence then the database-assisted identification of binding sites for proteins or protein classes and subsequent experimental verification might establish functionally relevant sites within this sequence. The binding transcription factors and interacting factors might already be present in the database.
Reinhard Hehl* Institut für Genetik, Technische Universität Braunschweig, Spielmannstr. 7, D-38106 Braunschweig, Germany. *e-mail:
[email protected] Edgar Wingender Research Group Bioinformatics, German Centre for Biotechnology, and BIOBASE GmbH, Mascheroder Weg 1, D-38124 Braunschweig, Germany.
Since the establishment of the central dogma of molecular biology, it has become obvious that the ‘transformation’ processes of information flow from DNA to RNA to protein are subject to a variety of control mechanisms. The key transmitting step that initiates the information flow, transcription, is mediated in eukaryotes by three different polymerases that transcribe distinct sets of genes. RNA polymerase II transcribes all genes that encode proteins. Many of these genes have a ‘core promoter’ comprising the initiator site (around +1, the position of the first transcribed nucleotide) and a TATA box at around −30. The transcription apparatus for RNA polymerase II is assembled at the core promoter and, in addition to the enzyme, this machinery includes several general transcription factors. TBP (TATAbinding protein), a subunit of transcription factor IID,
251
15
16
17
18
19
20
maize germination on cell cycle proteins. Physiol. Plant. 102, 573–581 Ball, K.L. and Lane, D.P. (1996) Human and plant proliferating-cell nuclear antigen have a highly conserved binding site for the p53-inducible gene product p21WAF1. Eur. J. Biochem. 237, 854–861 Matsumoto, T. et al. (1994) Expression of functional proliferating-cell nuclear antigen from rice (Oryza sativa) in Escherichia coli. Activity in association with human DNA polymerase δ. Eur. J. Biochem. 223, 179–187 den Boer, B.G.W. and Murray, J.A.H. (2000) Triggering the cell cycle in plants. Trends Cell Biol. 10, 245–250 Georgieva, E.I. et al. (1994) Maize embryos germination I. Cell cycle analysis. Planta 192, 118–124 Murray, J.A.H. et al. (1998) Plant D cyclins and retinoblastoma protein homologues. In Plant Cell Division (Francis, D. et al., eds), pp. 99–127, Portland Press, London, UK Walbot, V. (2000) A green chapter in the book of life. Nature 408, 794–795
is the primary DNA-recognition component for most promoters1. The presence of two types of TATAbinding proteins (TBPs) in plants suggests that, although transcription in eukaryotes is highly conserved, fundamental differences might exist2,3. The efficiency of the transcription initiation complex formation is largely influenced by the regulatory transcription factors that bind to short sequence elements that activate or repress genes in a manner that is specific for the tissue, the developmental stage or the stress conditions. These regulatory transcription factors interact with the general transcription factors directly or via coactivators4,5. To satisfy their specific biological requirements, plants have evolved unique regulatory mechanisms, involving completely new transcription factors that have yet to be found in animals. For example the WRKY (‘worky’) family of transcription factors, with probably up to 100 members in Arabidopsis, regulates the expression of a variety of target genes involved in the response to pathogen infection and other stresses6. Another plant-specific family of transcription factors is the Dof proteins, whose actions are related to biological processes unique to plants7. Dof proteins might contribute to the expression of genes involved in photosynthesis, in the response to stress and hormone signals, and in carbon metabolism8. Other examples of plant-specific factors include the homeodomain-ZIP (HD-ZIP) and GT-box-binding factors9. If a particular binding site occurs within a promoter, the relevant transcription factor can bind to this site, assuming that it is present in the nucleus and is in a competent state for binding. Such competent states can involve heterodimer formation or specific post-translational modifications. Transcription factors normally regulate more than one gene. The presence of a particular transcription factor binding site within
http://plants.trends.com 1360-1385/01/$ – see front matter © 2001 Elsevier Science Ltd. All rights reserved. PII: S1360-1385(01)01954-9
252
Opinion
TRENDS in Plant Science Vol.6 No.6 June 2001
Box 1. The history of the transcription factor database TRANSFAC Model data about transcriptional regulation has been collected since 1987, mainly for vertebrates, fungi and insects. The basis for transcriptional regulation is the recognition of short sequence elements by transcription factors, which translate regulatory genomic information into biological reality. Therefore, the basal structure of the first data collection comprised two tables: SITES and FACTORSa. This database was subsequently called TRANSFAC and, from it, a flat file version was constructedb. It was later transformed into a relational database systemc,d. At the same time, the flat-file version of the database was made available on the World-Wide Web (http://www.gene-regulation.de/)*. It was part of the concept to develop this and other databases, not only for encyclopedic purposes, but also to make them operational. For instance, using these database contents for the characterization and identification of individual regulatory elementsb. This led to the development of tools such as MatInspector (Ref. e) and PatSearch (Ref. f), which are now being replaced by Match and Patch. More recently, efforts have been made to increase the number of plant-specific data setsg,h. References a Wingender, E. (1988) Compilation of transcription regulating proteins. Nucleic Acids Res. 16, 1879–1902 b Wingender, E. et al. (1991) Regulatory DNA sequences: predictability of their function. In Genome Analysis – From Sequence to Function (BioTechForum Advances in Molecular Genetics) (Vol. 4) (Collins, J. and Driesel, A.J., eds), pp. 95–108, Hüthig, Heidelberg c Knuppel, R. et al. (1994) TRANSFAC retrieval program: a network model database of eukaryotic transcription regulating sequences and proteins. J. Comput. Biol. 1, 191–198 d Wingender, E. et al. (1996) TRANSFAC: a database on transcription factors and their DNA binding sites. Nucleic Acids Res. 24, 238–241 e Quandt, K. et al. (1995) MatInd and MatInspector: new fast and versatile tools for detection of consensus matches in nucleotide sequence data. Nucleic Acids Res. 23, 4878–4884 f Wingender, E. et al. (1997) TRANSFAC database as a bridge between sequence data libraries and biological function. In Pacific Symposium on Biocomputing ’97 (PSB’97) (Altman, R.B. et al., eds.), pp. 477–485, World Scientific g Wingender, E. et al. (2001) The TRANSFAC system on gene expression regulation. Nucleic Acids Res. 29, 281–283 h Wingender, E. et al. (2000) TRANSFAC: an integrated system for gene expression regulation. Nucleic Acids Res. 28, 316–319 *Most of the database tools described in this contribution are freely available to users from non-profit organizations. Users from commercial organizations are kindly requested to license the professional database version(s).
the regulatory regions of a set of genes can reflect their mode of regulation. Often, combinations of binding sites are responsible for regulated gene expression. The joint presence of two types of cis regulatory sequences is strong evidence for similar gene regulation, as recently shown for activated T cells10. In plants, combinatorial elements include those that control anthocyanin synthesis and abscisic-acidinduced gene expression11.
417 different names of plant transcription sites describing more than 159 plant promoters13. TRANSFAC (http://www.gene-regulation.de/), the first transcription factor database of eukaryotic cis-acting regulatory elements and trans-acting factors (Box 1), covers transcription factors from yeast to humans. Only TRANSFAC provides structural, expression, and functional information about the transcription factors. The number of plant transcription factors in the database has risen from 266 to 489 within the past year, and this was accompanied by a similar increase in the number of annotated sites14,15. To date, the number of sites in known plant promoters is 179, although this does not include artificial binding sites or binding sites derived from consensus sequences. There are 12 plantspecific MATRIX tables that represent 401 artificial binding sites. All plant species for which factors or sites are known are covered by the database. Once a relevant functional region has been delineated in a promoter, the sequence can be screened by using the Patch, SignalScan and Match software for real and putative transcription factor binding sites in the TRANSFAC database (Fig. 1). Once binding sites are discovered, they can be mutated specifically to test the involvement of the respective transcription factors in gene regulation (Fig. 1). Reconstruction experiments shed further light on the question of whether this site is responsible, either alone or in concert with other factors, for the observed gene expression. However, in reality, things can be much more complex. For example, database-assisted promoter analysis can identify one factor that is common to all expressed promoters (Fig. 2a). Because the binding site is distributed over a variable distance from the transcription start site, reconstruction experiments reveal that this site alone confers gene expression in a distance-independent mode. In a distance-dependent regulatory site, the binding sites are also found in genes that are not expressed (Fig. 2b). However, the genes that are expressed have the binding site at a specific distance from the transcription start site. These rather simplistic examples of promoter architecture might not represent the biological reality and composite elements might be responsible for specific expression profiles in most genes (Fig. 2c). Because the highly conserved core sequences of binding sites of transcription factors are relatively short, these sequences occur at a statistically predictable frequency in any given sequence. Therefore, it is helpful to have some experimental evidence about regulatory sequences before using database-assisted analysis.
Database-assisted plant promoter analysis
Currently, there are three databases that identify transcription factor binding sites or cis-acting sequences in plant promoters. PLACE (http://www.dna.affrc.go.jp/htdocs/PLACE/) is a database of 319 cis-acting regulatory DNA elements that were collected from previously published reports12. PlantCARE (http://sphinx.rug.ac.be:8080/ PlantCARE/index.htm) is a referential database with http://plants.trends.com
TRANSFAC and transcriptional regulation
TRANSFAC is a database of transcription factors and their genomic binding sites and DNA-binding profiles. This information resource is maintained as a relational database comprising ~100 tables, which, in the flat-file version, are condensed to six text-based files. The following description refers to the main components and contents of this flat-file system.
Opinion
Fig. 1. Identification of regulatory sites within a functionally delineated promoter sequence. The black lines show a putative plant promoter region that is sufficient for gene expression (+) when linked to a minimal promoter element (TATAbox) and a reporter gene. Database-assisted analysis identified four putative transcription factor binding sites (first line). The cis sequences bound by these factors are shown with different shades of blue in the promoter region; the transcription factors are displayed in red, green or yellow above the promoter fragment. Mutations that abolish factor binding (white boxes) identified one binding site as relevant for gene expression (−). This binding site alone is sufficient for gene expression, as shown by a reconstruction experiment involving the factor binding site alone.
TRENDS in Plant Science Vol.6 No.6 June 2001
Expression +
+
+
Factor:
253
other. In future, these protein–protein interactions will be classified as well. For instance, homo- or heterodimerization represents the most fundamental kind of interprotein complexing between transcription factors. At a higher level of complexity, this includes interactions between distinct transcription factors that are bound to distinct genomic sites and cause synergistic or antagonistic effects, which are typical for composite elements. Those protein–protein interactions that mediate the effect of the upstream transcription factor on the basal transcription initiation complex can constitute a third level of interaction.
–
CLASS
Most transcription factors have been hierarchically classified according to the properties of their DNAbinding domains. This classification can also be used to browse those transcription factors that have been assigned a location in the classification scheme. The CLASS table gives detailed explanations about individual classes and superclasses in this classification scheme.
+
+
TRENDS in Plant Science
SITE Accessing and using TRANSFAC
Users of the public version of the database need to register on the home page (http://www.generegulation.de/). From this home page, users have access to different areas such as databases, programs, papers, commercial offers and events. When going to the databases, they find a list of databases, of which TRANSFAC is the most relevant to plant scientists. On the same page, there are options to search TRANSFAC for the presence of a particular factor, gene, matrix or site. Furthermore, one can read detailed descriptions about the latest changes and the contents of each field in the database. There are 13 different featured programs on the home page. Currently, Patch, SignalScan and Match are the most relevant for identifying cis-acting sequences and binding factors in a given DNA sequence. FACTOR
The FACTOR table holds data about individual transcription factors. Information is given about synonyms, physicochemical, local and global structural, functional properties, amino acid sequence, and gene expression patterns of transcription factors. Many transcription factors have several splice variants, which can differ greatly in their functional properties, such as transcriptional activation or DNA binding, and can even act as antagonists to each other. All these variants are included in this table as individual entries. Protein–protein interactions are an important feature controlling the activity of transcription factors. For this reason, all those factors that are known to interact physically are linked with each http://plants.trends.com
The known genomic binding sites and sequences of DNA-binding transcription factors are given in the SITE table. The corresponding gene and the position of the binding site relative to its transcription start site (or another reference position, if appropriate) are given. The factor–site interactions are qualified by a number that reflects the confidence level of the underlying experimental evidence. Thus, merely suggested binding sites are rated 5, whereas a clear-cut identification of a transcription factor binding to a site complemented by functional evidence is rated 1. The genomic binding sites are linked to the corresponding entry in the GENE table. Artificial binding sequences are also documented in the SITE table; these might have been published in random selection studies to determine the DNA-binding properties of a certain transcription factor. Artificial binding sequences frequently serve as training sets to deduce a positional weight matrix. Finally, the SITE table also contains consensus strings using the 15-letter IUPAC alphabet; many of these have been published16. GENE
The GENE table contains the names and acronyms of all genes with at least one transcription factor binding site listed in TRANSFAC. The location of the individual binding sites is visualized. MATRIX
Once several binding sites for a given transcription factor are documented in the SITE table, they are aligned and a positional weight matrix is derived from them and stored in the MATRIX table. The lowest quality of binding sequences in the training set defines the quality of the matrix itself. Each matrix comes with information about
254
Opinion
(a)
TRENDS in Plant Science Vol.6 No.6 June 2001
Expression + + + – Factor: – – + + +
(b) + + + – – – – + –
Factor:
cldb/indexes.html), and GENE to COMPEL (http://compel.biomednet.nsc.ru), TRRD (http://wwwmgs.biomednet.nsc.ru/mgs/dbases/trrd4/) and soon GeneCard. Resources for plant-specific entries in TRANSFAC
When the TRANSFAC database was originally created, the emphasis was on vertebrates, fungi and insects; the number of plant entries has only increased recently (Box 1). The main tables with plant entries are FACTOR, SITE, MATRIX and GENE. All information that is entered into the databases is collected from published papers that are cited in the respective table. The main source of published data on transcription factors is the PubMed database (http://www.ncbi.nlm.nih.gov/PubMed/) and also resources such as the AGRICOLA database (http://www.nal.usda.gov/ag98/) and the CD-ROM version of the Biological Abstracts. Connected tools
(c) + + + – – – – – +
Factor:
TRENDS in Plant Science
Fig. 2. Identification of functionally relevant sites within similarly regulated promoters. Three examples of regulatory sites within similar expressed genes are shown. The black lines show a putative plant promoter region that is sufficient for gene expression (+) when linked to a minimal promoter element (TATA-box) and a reporter gene. Database-assisted analysis identified four putative transcription factor binding sites. The cis sequences bound by these factors are shown with different shades of blue in the promoter region; the transcription factors are displayed in red, green or yellow above the promoter fragment. The first six lines represent promoters and their state of expression (+/−). The last three lines show reconstruction experiments based on the results of the database-assisted analysis that identified common factor binding sites. (a). A single site is responsible in this case for gene expression, in a distance-independent manner. (b) In this case, a single site acts in a distance-dependent fashion. (c) A composite element is responsible for gene expression in this case.
its statistical basis and the experimental set up used to obtain the training sequences. All the underlying sequences, if available, are deposited in the SITE table, and proper links are given that also include parameters for the corresponding alignment. Database links
Wherever appropriate, links to other databases are included. Thus, FACTOR entries might be linked to the EMBL data library, with SwissProt (Ref. 17), PIR (Ref. 18) and PDB (Ref. 19). SITE data are connected with EMBL (Ref. 20) and EPD (Ref. 21). CLASS contains links to ProSite (Ref. 22), CELL to CLDB (Cell Line Database; http://www.biotech.ist.unige.it/ http://plants.trends.com
The sequence data and the matrices of the TRANSFAC database can be used to analyse the occurrence of potential transcription factor binding sites, using for instance, the Patch and SignalScan programs23. The whole set of sequence elements has been split into several libraries for fungal, insect, plant or vertebrate sites. The list of consensus sequences and the sequence information stored in TRRD are also provided as separate libraries24. When using Patch on the gene-regulation home page, there is a field for entering the DNA sequence as well as a link to a training set of sequence examples. After entering the sequence, there is the option to restrict the search to plant sites and to change specific parameters of the query. Submitting the query provides a list of all sites and their locations in the submitted sequence. Each site is linked to its SITE entry, which gives the available data about this site including, for example, the name, where it was originally found (gene promoter or artificial sequence) and the position relative to either transcription or translation start. Furthermore, there are direct links to any known factors that bind to this site and to the literature where the data was published. The SignalScan and Match programs can be used in a similar way. The positional weight matrices contained in the MATRIX table are used by the programs MatInspector (Ref. 25) and Match. For this purpose, all matrices were evaluated in systematically determining false negative and false positive hits under all thresholds26. As a result, a list of individually optimized thresholds for most search patterns has been obtained. The same matrix library is also used by the FastM program27, which scans DNA sequences for the occurrence of userdefined combinations of several transcription factor binding sites. Some of these combinations might reveal properties of ‘composite elements’as they are systematically stored in the database COMPEL (Ref. 28). However, the COMPEL database currently
Opinion
Acknowledgements Our work was supported in part by the German Ministry of Education and Research (BMBF), by a grant from the European Commission and by a grant from the Forschungsschwerpunkt Agrarbiotechnologie des Landes Niedersachsen. We are also grateful to Norbert Käufer for helpful suggestions on the manuscript.
TRENDS in Plant Science Vol.6 No.6 June 2001
has no entries for plant-specific composite elements. It has been proved in several mammalian examples that only context-sensitive analyses for combinations of transcription factor binding sites can lead to the required specificity in the in silico identification of regulatory genomic regions10,29,30 and the same principle will probably apply to plant regulatory regions as well11. Furthermore, additional software tools such as Motif Sampler (http://www.esat. kuleuven.ac.be/~thijs/Work/MotifSampler.html) can be used to identify common motifs in similarly regulated promoters before searching the TRANSFAC database for binding sites of transcription factors in these motifs. Although the databases provide only experimental data, they should be supplemented by analogously structured data resources on in silico annotated regulatory elements30. It should then be possible to extend the combinatoric analysis by combining experimentally proved and in silico annotated regulatory elements to gain insight in putative composite element structures. From analysis to design
Although there is still much to do before all the published data on plant transcription factors are represented in the TRANSFAC database, we can speculate about the potential use of this database in
References 1 Burley, S.K. and Roeder, R.G. (1996) Biochemistry and structural biology of transcription factor IID (TFIID). Annu. Rev. Biochem. 65, 769–799 2 Heard, D.J. et al. (1993) Both Arabidopsis TATA binding protein (TBP) isoforms are functionally identical in RNA polymerase II and III transcription in plant cells: evidence for genespecific changes in DNA binding specificity of TBP. EMBO J. 12, 3519–3528 3 Vogel, J.M. et al. (1993) Expression of the two maize TATA binding protein genes and function of the encoded TBP proteins by complementation in yeast. Plant Cell 5, 1627–1638 4 Washburn, K.B. et al. (1997) Coactivators and TAFs of transcription activation in wheat. Plant Mol. Biol. 35, 1037–1043 5 Le Gourrierec, J. et al. (1999) Transcriptional activation by Arabidopsis GT-1 may be through interaction with TFIIA–TBP–TATA complex. Plant J. 18, 663–668 6 Eulgem, T. et al. (2000) The WRKY superfamily of plant transcription factors. Trends Plant Sci. 5, 199–206 7 Yanagisawa, S. (1996) Dof DNA binding proteins contain a novel zinc finger motif. Trends Plant Sci. 1, 213–214 8 Yanagisawa, S. (2000) Dof1 and Dof2 transcription factors are associated with expression of multiple genes involved in carbon metabolism in maize. Plant J. 21, 281–288 9 Meisel, L. and Lam, E. (1997) Switching of gene expression: analysis of the factors that spatially and temporally regulate plant gene expression. Genet. Eng. 19, 183–199 10 Kel, A. et al. (1999) Recognition of NFATp/AP-1 composite elements within genes induced upon the activation of immune cells. J. Mol. Biol. 288, 353–376 http://plants.trends.com
255
the future. The accuracy with which particular transcription factor binding sites will be discovered in a given sequence will largely depend on the available experimental data entered into the database. With the advent of DNA chip technology, we might soon know the expression pattern of every gene in model organisms. Because this will not only involve putative target genes but also the respective transcription factors, we might eventually know all the cis and trans factors that regulate the expression of every gene. It is also important to extend the work from the transcriptome31 to the proteome32 because posttranslational modifications add another level of complexity to the analysis of the regulation of gene expression. This was exemplified recently by the targeted destabilization of the bZIP transcription factor HY5 during light-regulated development33. Furthermore, the non-cell-autonomous activity of transcription factors and the location of cis regulatory elements downstream of the transcription start site need to be considered34,35. If the databases are completed, we might be able to apply or generate novel tools to design custom-made plant promoters in silico that direct the expression of a gene in time and space. This can be used to answer basic scientific questions and might also lead to transgenic crop plants in which the gene product is only expressed at the desired time and in the desired tissue.
11 Singh, K.B. (1998) Transcriptional regulation in plants: the importance of combinatorial control. Plant Physiol. 118, 1111–1120 12 Higo, K. et al. (1999) Plant cis-acting regulatory DNA elements (PLACE) database: 1999. Nucleic Acids Res. 27, 297–300 13 Rombauts, S. et al. (1999) PlantCARE, a plant cisacting regulatory element database. Nucleic Acids Res. 27, 295–296 14 Wingender, E. et al. (2001) The TRANSFAC system on gene expression regulation. Nucleic Acids Res. 29, 281–283 15 Wingender, E. et al. (2000) TRANSFAC: an integrated system for gene expression regulation. Nucleic Acids Res. 28, 316–319 16 Faisst, S. and Meyer, S. (1992) Compilation of vertebrate-encoded transcription factors. Nucleic Acids Res. 20, 3–26 17 Bairoch, A. and Apweiler, R. (2000) The SWISSPROT protein sequence database and its supplement TrEMBL in 2000. Nucleic Acids Res. 28, 45–48 18 Barker, W.C. et al. (2000) The protein information resource (PIR). Nucleic Acids Res. 28, 41–44 19 Berman, H.M. et al. (2000) The Protein Data Bank. Nucleic Acids Res. 28, 235–242 20 Baker, W. et al. (2000) The EMBL nucleotide sequence database. Nucleic Acids Res. 28, 19–23 21 Perier, R.C. et al. (2000) The eukaryotic promoter database (EPD). Nucleic Acids Res. 28, 302–303 22 Hofmann, K. et al. (1999) The PROSITE database, its status in 1999. Nucleic Acids Res. 27, 215–219 23 Prestridge, D.S. (1991) SIGNAL SCAN: a computer program that scans DNA sequences for eukaryotic transcriptional elements. CABIOS 7, 203–206. 24 Kolchanov, N.A. et al. (2000) Transcription regulatory regions database (TRRD): its status in 2000. Nucleic Acids Res. 28, 298–301
25 Quandt, K. et al. (1995) MatInd and MatInspector: new fast and versatile tools for detection of consensus matches in nucleotide sequence data. Nucleic Acids Res. 23, 4878–4884 26 Pickert, L. et al. (1998) Transcription regulatory region analysis using signal detection and fuzzy clustering. Bioinformatics 14, 244–251 27 Klingenhoff, A. et al. (1999) Functional promoter modules can be detected by formal models independent of overall nucleotide sequence similarity. Bioinformatics 15, 180–186 28 Kel, O.V. et al. (1995) A compilation of composite regulatory elements affecting gene transcription in vertebrates. Nucleic Acids Res. 23, 4097–4103 29 Frech, K. et al. (1998) Muscle actin genes: a first step towards computational classification of tissue specific promoters. In Silico Biol. 1, 0005 (http://www.bioinfo.de/isb/1998/01/0005/) 30 Maleck, K. et al. (2000) The transcriptome of Arabidopsis thaliana during systemic acquired resistance. Nat. Genet. 26, 403–410 31 Velculescu, V.E. et al. (1997) Characterization of the yeast transcriptome. Cell 88, 243–251 32 Lottspeich, F. (1999) Proteome analysis: a pathway to the functional analysis of proteins. Angew. Chem., Int. Ed. Engl. 38, 2476–2492 33 Osterlund, M.T. et al. (2000) Targeted destabilization of HY5 during light-regulated development of Arabidopsis. Nature 405, 462–466 34 Sieburth, L.E. and Meyerowitz, E.M. (1997) Molecular dissection of the AGAMOUS control region shows that cis elements for spatial regulation are located intragenically. Plant Cell 9, 355–365 35 Sessions, A. et al. (2000) Cell–cell signaling and movement by the floral transcription factors LEAFY and APETALA1. Science 289, 779–782
