## INTRODUCTION

The quadruplex structures formed by guanine-rich nucleic acid sequences have received significant attention recently because of increasing evidence for their role in important biological processes and as therapeutic targets ( 15 ). The G-quadruplex structure, also known as a G-quartet, is formed by repeated folding of either the single polynucleotide molecule or by association of two or four molecules. The structure consists of stacked G-tetrads, which are square co-planar arrays of four guanine bases each ( 6 ). G-quadruplex is stabilized with cyclic Hoogsteen hydrogen bonding between the four guanines within each tetrad. The present work focuses only on the unimolecular quadruplexes, since it is more likely to be encountered in physiological conditions ( 7 , 8 ).

Guanine-rich sequences capable of forming G-quadruplexes are found in telomeres, promoter regions, transcribed and other biologically important regions of the mammalian genomes. G-quadruplex DNA has been suggested to regulate DNA replication in retinoblastoma susceptibility gene (Rb) region ( 9 ). This structure may control cellular proliferation at telomeric level and by transcriptional regulation of oncogenes like c-myc ( 2 , 10 , 11 ) and c-kit ( 12 ). Formation of G-quadruplex seems to be regulated through interactions with cellular proteins. While some proteins help stabilize the structure ( 13 ), others are known to resolve it ( 1 , 4 , 14 , 15 ). Proteins and chemicals that stabilize the G-quadruplex structure can inhibit telomerase action and, therefore, are being evaluated as anticancer therapeutic agents ( 1620 ). Chemical compounds that inhibit G-quadruplex helicase activity may also be capable of regulating cellular proliferation ( 4 ). G-quadruplexes are also being eyed as potential antimicrobial agents due to their ability to transport monovalent anions ( 21 ).

G-quadruplex motifs in the RNA have been shown to play significant roles in mRNA turnover ( 1 ), FMRP binding ( 22 ), translation initiation ( 23 ) as well as repression ( 24 ). We have shown previously that a G-rich sequence (GRS) can mediate 3′ end processing of mammalian pre-mRNAs by interacting with DSEF-1/hnRNPH/H′ protein ( 2527 ). Members of the hnRNP H protein family recognize G-rich motifs capable of forming G-quadruplexes and are known to regulate polyadenylation and splicing events in mammalian transcripts ( 2830 ). Regulated RNA processing is an essential component of differential gene expression which is central to many important biologically processes. More than half of human genes are known to have alternative polyadenylation ( 31 ). Over two-thirds of human genes are thought to undergo alternative splicing ( 32 ). Sequences capable of forming G-quadruplexes found in the vicinity of polyadenylation and splice sites act as regulators by interacting with hnRNP F and H proteins ( 2527 , 30 , 33 ). Whether quadruplex structure directly plays a role in regulating RNA processing events requires investigation.

## METHODS

### QGRS definition

The main goal of the QGRS Mapper program is to predict the presence of QGRS in nucleotide entries. These putative G-quadruplexes are identified using the following motif.

${\mathbf{\hbox{ G }}}_{x}{\mathbf{\hbox{ N }}}_{{y}_{1}}{\mathbf{\hbox{ G }}}_{x}{\mathbf{\hbox{ N }}}_{{y}_{2}}{\mathbf{\hbox{ G }}}_{x}{\mathbf{\hbox{ N }}}_{{y}_{3}}{\mathbf{\hbox{ G }}}_{x}$
Here x = number of guanine tetrads in the G-quadruplex and y1 , y2 , y3 = length of gaps (i.e. the length of the loops connecting the guanine tetrads). The motif consists of four equal length sets of guanines (which we call G-groups), separated by arbitrary nucleotide sequences, with the following restrictions.

• The sequence must contain at least two tetrads (i.e. x2 ). Although structures with three or more G-tetrads are considered to be more stable, many nucleotide sequences are known to form quadruplexes with two G-tetrads ( 37 , 38 ). QGRS Mapper is meant to be a flexible and comprehensive tool for investigating G-quadruplexes; hence it considers sequences with two tetrads.

• By default, only QGRS of maximum length of 30 bases are considered. However, the program gives the user the option to search for sequences up to 45 bases. This restriction on the length of the sequences being considered is in agreement with recent literature ( 34 , 35 ). The maximum length of 30 bases restricts G-groups to a maximum size of 6.

• The gaps or loops between the G-groups may be arbitrary in composition or length (within the overall restrictions on the length of QGRS). The program gives the user the option to search for QGRS having loops with a specified length range (e.g. the user can search for QGRS with loops of lengths between 1 and 4). The user can also specify a string that one or more loops of each QGRS must contain. This string can be given as a regular expression. For example, entering the regular expression ‘T{3,5}’ will search for QGRS having one or more loops that contain three to five consecutive T's.

• Also, at most one of the gaps is allowed to be of zero length

Table 1 shows some examples of valid QGRS. The guanine groups which form the tetrads are underlined.

The first sequence has four tetrads and equal length gaps. This would seem to provide a G-quadruplex that is the most stable of the three sequences. The second sequence is notable for the significant differences in the size of its loops. The third sequence has two tetrads, even though three of the G-groups could have included another G (since all G-groups must be equal in size).

### G-scores

We have devised a scoring system that evaluates a QGRS for its likelihood to form a stable G-quadruplex. Higher scoring sequences will make better candidates for G-quadruplexes. The scoring method uses the following principles which are based on previous studies ( 34 , 35 , 3942 ).

• Shorter loops are more common than longer loops.

• G-quadruplexes tend to have loops roughly equal in size.

• The greater the number of guanine tetrads, the more stable the quadruplex.

The computed G-scores are dependent on the user selected maximum QGRS length. The highest possible G-score, using the default maximum QGRS length of 30, is 105. Here is a sequence attaining that score:

• GGGGGG T GGGGGG T GGGGGG T GGGGGG .

### Eliminating QGRS overlaps

Two QGRS are said to overlap if their positions in the nucleotide sequence overlap. QGRS Mapper will start with a nucleotide sequence, find all QGRS occurring in the sequence and then produce a non-overlapping set of QGRS. Overlaps are eliminated by selecting the higher scoring QGRS. In the non-overlapping view, only this sequence will be displayed, although the user can request that all overlapping sequences be displayed.

Please see supplementary materials at NAR online for more details on the elimination of overlapping QGRS.

## FEATURES

### Design and implementation

QGRS Mapper is a web-based program, written in PHP, with Java being used for some of its graphics. The program takes a nucleotide sequence from NCBI (or as provided by the user) and analyzes it for the presence of putative G-quadruplexes. The structure of QGRS Mapper is summarized in Table 2 .

### Search and analysis

QGRS Mapper allows the user to search for putative G-quadruplexes in a variety of ways. It is possible to enter a nucleotide sequence in raw or FASTA format for analysis. One can search and analyze gene sequences by Gene ID, Gene name or symbol, accession number or GI number for an NCBI nucleotide sequence entry. The user can opt to change the maximum length of QGRS that will be searched for (the default maximum length being 30) and change the minimum sized G-group (which is two by default). Also, the user can specify that the loops in the QGRS fall within a given size range and that one or more loops of the QGRS contain a given string (for which the user may enter a regular expression). The web page for QGRS analysis can be seen in Figure 1 .

After entering a sequence in raw or FASTA format, QGRS Mapper will search the sequence for occurrences of QGRS. The user may enter any combination of the letters A, C, T, G, U, N.

The Gene ID field allows the user to search the NCBI Entrez Gene database. QGRS Mapper will connect to NCBI, download and parse the gene entry, and then analyze the transcribed region of its nucleotide sequence for the presence of QGRS. For example, entering the gene ID 403437 results in downloading the Brca1 gene sequence for Canis familiaris . Using the default QGRS search parameters, QGRS Mapper finds 156 non-overlapping QGRS and 3394 overlapping QGRS in the transcribed region of this gene.

The Gene Name or Gene Symbol field also allows the user to search the NCBI databases for all such genes. Entering the gene name Bcl2 results in nine different hits which are displayed in Table 3 .

All nine of these entries can be analyzed for the occurrence of QGRS. Clicking on the Gene ID takes the user to the respective Entrez Gene entry. Clicking on the last column initiates analysis of the selection by QGRS Mapper.

Similarly, the user can also enter an NCBI accession number to search for gene sequences. For example, searching the accession number AF312033 results in 12 hits being displayed for this GenBank nucleotide sequence entry.

The search phase of the program is followed by an analysis of the QGRS contained in the query sequence. In this phase of QGRS Mapper, the sequence data downloaded previously is analyzed to identify and map all QGRS relative to locations such as splice sites in exons/introns, and poly(A) site (if these locations are known). Furthermore the QGRS are scored by the method described above. The computed G-score is used to eliminate overlapping QGRS.

At times, QGRS Mapper must analyze a considerable amount of data. For example, the mouse version of the gene PTPRU, which is 69822 bases long, contains 94681 QGRS of length up to 45 bases. QGRS Mapper will find, analyze and map all of these sequences. During this analysis a message is displayed indicating the estimated time left to completion.

### QGRS Mapper output

After the analysis of overlaps is completed, QGRS Mapper displays a summary of its findings, in the Gene View. This summary includes basic gene information such as the gene ID, gene symbol, gene name, a link to the NCBI entry, organism name, chromosome number and number of products and poly(A) signals. Information is also given for each product, such as the number of exons and introns, number of QGRS (non-overlapping and overlapping), number of QGRS found near RNA processing sites, and a visual map of the product.

As an example, the Gene View for the human GREB1 is displayed in Figure 2 , showing the table of gene information and product information for the first product (the output for all products may be seen in the supplementary material).

At this stage in the analysis the user can choose among three further displays: ‘Data View’, ‘Data View (with overlaps)’ and ‘Graphics View’. This can be done for the entire gene or for any particular product.

In the Data View, a table is displayed showing information for each of the set of non-overlapping QGRS. This table displays the position of the QGRS, which exon/intron it appears in, its distance from 3′ and 5′ splice sites, the QGRS sequence (with each G-group underlined) and the corresponding G-score. Similar display is also shown for each QGRS mapped to poly(A) region in the product. If the user requests the Data View for the entire gene, then the QGRS information is shown for each product. The ‘Data View (with overlaps)’ gives the same information but shows the locations of all QGRS. Figure 3 shows the Data View for product 1 of the GREB1 gene.

The user can also choose the Graphics View to give a visual display of the location of QGRS. This allows the user to see the location of QGRS relative to exons and introns (if that information is available). The Graphics View has the following components.

• A graphic display of the entire gene (showing the location of the exons). This display includes a sliding window that can be used to focus on any particular segment of the gene. This window may be dragged to the left or right to change position within the gene.

• A magnified view of the fragment of the gene within the sliding window.

• A graph showing the location of QGRS within the fragment, with each QGRS being displayed by a bar whose height represents its G-score.

• A vertical slider that allows the user to change the size of the window. This allows the user to zoom in or out on any part of the gene. The sliding window on the gene expands or contracts as one zooms in or out. It is possible to see the nucleotide sequence of the product at maximum zoom levels.

The Graphics View for the entire gene shows the G-score graph together with an exon/intron map for each product. This allows the user to visually compare the location of QGRS for each product relative to that of splice sites. The Graphics View for the first product of the GREB1 gene is represented in Figure 4 . The Graphics View for the entire GREB1 gene may be seen in the Supplementary Data.

## CONCLUSIONS

QGRS Mapper is a user-friendly web-based server that provides computational tools for prediction of Quadruplex forming G-rich sequences in the nucleotide sequences identified or provided by the user. The program offers many options, including user-defined composition of the quadruplex. It can analyze DNA or RNA sequence provided by the user in the raw or FASTA format. The application also provides tools for searching and retrieving gene/nucleotide entries from a variety of NCBI databases. There are several options for data output format, including an interactive graphic module.

Researchers interested in evaluating the ability of nucleotide sequences to form unimolecular G-quadruplexes will find QGRS Mapper to be very useful. Owing to the flexible and comprehensive nature of the design, it is expected to serve a variety of scientists. The application will be especially attractive to individuals interested in exploring the role of G-quadruplexes in regulated RNA processing. We are using the server to perform a large-scale analysis of alternatively processed mammalian transcripts. We are particularly interested in studying the composition and distribution patterns of G-quadruplexes in the transcribed regions of mammalian genes.

## SUPPLEMENTARY DATA

Supplementary Data are available at NAR online.

Table 1

QGRS examples (G-groups are underlined)

Sequence QGRS parameters
GGGG T GGGG T GGGG T GGGG x = 4 , y1 = 1 , y2 = 1 , y3 = 1
GGG T GGG TGGCAGAGCT GGG CT GGG x = 3 , y1 = 1 , y2 = 10 , y3 = 2
GG GC GG GCT GG GTT GG x = 2 , y1 = 2 , y2 = 3 , y3 = 3
Sequence QGRS parameters
GGGG T GGGG T GGGG T GGGG x = 4 , y1 = 1 , y2 = 1 , y3 = 1
GGG T GGG TGGCAGAGCT GGG CT GGG x = 3 , y1 = 1 , y2 = 10 , y3 = 2
GG GC GG GCT GG GTT GG x = 2 , y1 = 2 , y2 = 3 , y3 = 3

Table 2

QGRS Mapper

Component Function
Retrieve Query, retrieve and parse gene information from NCBI or obtain nucleotide sequence from user.
Search Identify all QGRS in a given sequence.
Map QGRS are mapped relative to locations such as splice sites in introns/exons, and poly(A) site. If original sequence was provided by user in raw or FASTA format, this step is omitted.
Score QGRS are assigned scores related to their potential to form G-quadruplexes
Overlap Overlapping QGRS are eliminated using a greedy algorithm
Results Display summary of QGRS analysis in three forms: Gene View, Data View and Graphics View
Component Function
Retrieve Query, retrieve and parse gene information from NCBI or obtain nucleotide sequence from user.
Search Identify all QGRS in a given sequence.
Map QGRS are mapped relative to locations such as splice sites in introns/exons, and poly(A) site. If original sequence was provided by user in raw or FASTA format, this step is omitted.
Score QGRS are assigned scores related to their potential to form G-quadruplexes
Overlap Overlapping QGRS are eliminated using a greedy algorithm
Results Display summary of QGRS analysis in three forms: Gene View, Data View and Graphics View

Figure 1

QGRS Mapper Analysis Page. This page allows the user to enter a nucleotide sequence or to search NCBI for a nucleotide entry that will then be analyzed for the presence of QGRS. The user may enter the sequence in raw or FASTA format or search NCBI by Gene ID, Gene name or Accession number. There are options to define the QGRS for the search, including: the maximum length, the minimum number of guanine tetrads, loop length and composition.

Figure 1

QGRS Mapper Analysis Page. This page allows the user to enter a nucleotide sequence or to search NCBI for a nucleotide entry that will then be analyzed for the presence of QGRS. The user may enter the sequence in raw or FASTA format or search NCBI by Gene ID, Gene name or Accession number. There are options to define the QGRS for the search, including: the maximum length, the minimum number of guanine tetrads, loop length and composition.

Table 3

Search Results for Bcl2. (Gene ID links to Entrez Gene)

No. Gene symbol Gene ID Organism Analyze for QGRS
Bcl2 12043 Mus musculus GO
bcl2 570772 Danio rerio GO
BCL2 403416 C.familiaris GO
BCL2 596 Homo sapiens GO
BCL2 281020 Bos taurus GO
BCL2 494190 Ovis aries GO
BCL2 396282 Gallus gallus GO
Bcl2 24224 Rattus norvegicus GO
BCL2 493934 Felis catus GO
No. Gene symbol Gene ID Organism Analyze for QGRS
Bcl2 12043 Mus musculus GO
bcl2 570772 Danio rerio GO
BCL2 403416 C.familiaris GO
BCL2 596 Homo sapiens GO
BCL2 281020 Bos taurus GO
BCL2 494190 Ovis aries GO
BCL2 396282 Gallus gallus GO
Bcl2 24224 Rattus norvegicus GO
BCL2 493934 Felis catus GO

Figure 2

GREB1—Gene View. The first table in this view gives a summary of the gene being analyzed. This includes information such as the gene ID, name, symbol, a link to the corresponding NCBI entry, and the number of alternatively processed products and poly(A) signals found in the gene. The total number of QGRS in the gene is given for both non-overlapping and overlapping sequences. For each product, the numbers of exons and introns that were analyzed are given, together with the number of QGRS. Also, an exon/intron map is given. For the entire gene or for a particular product the user may select from three buttons to obtain more detailed information about the QGRS in the gene. These buttons are labeled Data View, Data View (with overlaps) and Graphics View. Only one product is shown in the figure. The screenshot showing all products can be found in the Supplementary Data.

Figure 2

GREB1—Gene View. The first table in this view gives a summary of the gene being analyzed. This includes information such as the gene ID, name, symbol, a link to the corresponding NCBI entry, and the number of alternatively processed products and poly(A) signals found in the gene. The total number of QGRS in the gene is given for both non-overlapping and overlapping sequences. For each product, the numbers of exons and introns that were analyzed are given, together with the number of QGRS. Also, an exon/intron map is given. For the entire gene or for a particular product the user may select from three buttons to obtain more detailed information about the QGRS in the gene. These buttons are labeled Data View, Data View (with overlaps) and Graphics View. Only one product is shown in the figure. The screenshot showing all products can be found in the Supplementary Data.

Figure 3

GREB1—Data View. A table of gene information is given at the top of the figure. The second table lists all QGRS mapped to exons and introns of the product, including information about the position of the QGRS, its distance from 3′ and 5′ splice sites, the actual sequence (underlining the G-groups) and its G-score.

Figure 3

GREB1—Data View. A table of gene information is given at the top of the figure. The second table lists all QGRS mapped to exons and introns of the product, including information about the position of the QGRS, its distance from 3′ and 5′ splice sites, the actual sequence (underlining the G-groups) and its G-score.

Figure 4

GREB1—Graphics View. A graphic representation of the gene, a sliding window that shows a fragment of the gene, and a graph showing the location and G-score for the non-overlapping QGRS in that fragment. A tool is provided to let the user zoom in or out on any portion of the gene. The nucleotide sequence is visible at maximum zoom levels.

Figure 4

GREB1—Graphics View. A graphic representation of the gene, a sliding window that shows a fragment of the gene, and a graph showing the location and G-score for the non-overlapping QGRS in that fragment. A tool is provided to let the user zoom in or out on any portion of the gene. The nucleotide sequence is visible at maximum zoom levels.

We thank Manuel Viotti for assistance with the testing of QGRS Mapper. This project was funded in part by a grant from the Ramapo College Foundation of Ramapo College of New Jersey. Funding to pay the Open Access publication charges for this article was provided by the Divisions of Student Affairs and Academic Affairs of Ramapo College of New Jersey.

Conflict of interest statement . None declared.

## REFERENCES

1
Simonsson, T.
2001
G-quadruplex DNA structures—variations on a theme
Biol. Chem
.
382
621
–628
2
Simonsson, T., Pecinka, P., Kubista, M.
1998
DNA tetraplex formation in the control region of c-myc
Nucleic Acids Res
.
26
1167
–1172
3
Davis, J.T.
2004
G-quartets 40 years later: from 5′-GMP to molecular biology and supramolecular chemistry
Angew Chem. Int. Ed. Engl
.
43
668
–698
4
Sharma, S., Doherty, K.M., Brosh, R.M., ,Jr.
2005
DNA helicases as targets for anti-cancer drugs
Curr. Med. Chem. Anticancer Agents

5
183
–199
5
Kelland, L.R.
2005
Overcoming the immortality of tumour cells by telomere and telomerase based cancer therapeutics—current status and future prospects
Eur. J. Cancer

41
971
–979
6
Gellert, M., Lipsett, M.N., Davies, D.R.
1962
Helix formation by guanylic acid

48
2013
–2018
7
Schaffitzel, C., Berger, I., Postberg, J., Hanes, J., Lipps, H.J., Pluckthun, A.
2001
In vitro generated antibodies specific for telomeric guanine-quadruplex DNA react with Stylonychia lemnae macronuclei

98
8572
–8577
8
Halder, K. and Chowdhury, S.
2005
Kinetic resolution of bimolecular hybridization versus intramolecular folding in nucleic acids by surface plasmon resonance: application to G-quadruplex/duplex competition in human c-myc promoter
Nucleic Acids Res
.
33
4466
–4474
9
Xu, Y. and Sugiyama, H.
2006
Formation of the G-quadruplex and i-motif structures in retinoblastoma susceptibility genes (Rb)
Nucleic Acids Res
.
34
949
–954
10
Phan, A.T., Modi, Y.S., Patel, D.J.
2004
Propeller-type parallel-stranded G-quadruplexes in the human c-myc promoter
J. Am. Chem. Soc
.
126
8710
–8716
11
Phan, A.T., Kuryavyi, V., Gaw, H.Y., Patel, D.J.
2005
Small-molecule interaction with a five-guanine-tract G-quadruplex structure from the human MYC promoter
Nature Chem. Biol
.
1
167
–173
12
Rankin, S., Reszka, A.P., Huppert, J., Zloh, M., Parkinson, G.N., Todd, A.K., Ladame, S., Balasubramanian, S., Neidle, S.
2005
Putative DNA quadruplex formation within the human c-kit oncogene
J. Am. Chem. Soc
.
127
10584
–10589
13
Paeschke, K., Simonsson, T., Postberg, J., Rhodes, D., Lipps, H.J.
2005
Telomere end-binding proteins control the formation of G-quadruplex DNA structures in vivo
Nature Struct. Mol. Biol
.
12
847
–854
14
Zaug, A.J., Podell, E.R., Cech, T.R.
2005
Human POT1 disrupts telomeric G-quadruplexes allowing telomerase extension in vitro

102
10864
–10869
15
Vaughn, J.P., Creacy, S.D., Routh, E.D., Joyner-Butt, C., Jenkins, G.S., Pauli, S., Nagamine, Y., Akman, S.A.
2005
The DEXH protein product of the DHX36 gene is the major source of tetramolecular quadruplex G4-DNA resolving activity in HeLa cell lysates
J. Biol. Chem
.
280
38117
–38120
16
Shin-Ya, K.
2005
Novel antitumor and neuroprotective substances discovered by characteristic screenings based on specific molecular targets
Biosci. Biotechnol. Biochem
.
69
867
–872
17
Tahara, H., Shin-Ya, K., Seimiya, H., Yamada, H., Tsuruo, T., Ide, T.
2005
G-quadruplex stabilization by telomestatin induces TRF2 protein dissociation from telomeres and anaphase bridge formation accompanied by loss of the 3′ telomeric overhang in cancer cells
Oncogene

25
1955
–1966
18
Zhou, J.M., Zhu, X.F., Lu, Y.J., Deng, R., Huang, Z.S., Mei, Y.P., Wang, Y., Huang, W.L., Liu, Z.C., Gu, L.Q., et al.
2006
Senescence and telomere shortening induced by novel potent G-quadruplex interactive agents, quindoline derivatives, in human cancer cell lines
Oncogene

25
503
–511
19
Moore, M.J., Schultes, C.M., Cuesta, J., Cuenca, F., Gunaratnam, M., Tanious, F.A., Wilson, W.D., Neidle, S.
2006
Trisubstituted acridines as G-quadruplex telomere targeting agents. Effects of extensions of the 3,6- and 9-side chains on quadruplex binding, telomerase activity, and cell proliferation
J. Med. Chem
.
49
582
–599
20
Ladame, S., Schouten, J.A., Roldan, J., Redman, J.E., Neidle, S., Balasubramanian, S.
2006
Exploring the recognition of quadruplex DNA by an engineered cys(2)-his(2) zinc finger protein
Biochemistry

45
1393
–1399
21
Kaucher, M.S., Harrell, W.A., Jr, Davis, J.T.
2006
A unimolecular G-quadruplex that functions as a synthetic transmembrane Na+ transporter
J. Am. Chem. Soc
.
128
38
–39
22
Bashkirov, V.I., Scherthan, H., Solinger, J.A., Buerstedde, J.M., Heyer, W.D.
1997
A mouse cytoplasmic exoribonuclease (mXRN1p) with preference for G4 tetraplex substrates
J. Cell Biol
.
136
761
–773
23
Bonnal, S., Schaeffer, C., Creancier, L., Clamens, S., Moine, H., Prats, A.C., Vagner, S.
2003
A single internal ribosome entry site containing a G quartet RNA structure drives fibroblast growth factor 2 gene expression at four alternative translation initiation codons
J. Biol. Chem
.
278
39330
–39336
24
Oliver, A.W., Bogdarina, I., Schroeder, E., Taylor, I.A., Kneale, G.G.
2000
Preferential binding of fd gene 5 protein to tetraplex nucleic acid structures
J. Mol. Biol
.
301
575
–584
25
Bagga, P.S., Ford, L.P., Chen, F., Wilusz, J.
1995
The G-rich auxiliary downstream element has distinct sequence and position requirements and mediates efficient 3′ end pre-mRNA processing through a trans-acting factor
Nucleic Acids Res
.
23
1625
–1631
26
Bagga, P.S., Arhin, G.K., Wilusz, J.
1998
DSEF-1 is a member of the hnRNP H family of RNA-binding proteins and stimulates pre-mRNA cleavage and polyadenylation in vitro
Nucleic Acids Res
.
26
5343
–5350
27
Arhin, G.K., Boots, M., Bagga, P.S., Milcarek, C., Wilusz, J.
2002
Downstream sequence elements with different affinities for the hnRNP H/H′ protein influence the processing efficiency of mammalian polyadenylation signals
Nucleic Acids Res
.
30
1842
–1850
28
Veraldi, K.L., Arhin, G.K., Martincic, K., Chung-Ganster, L.H., Wilusz, J., Milcarek, C.
2001
hnRNP F influences binding of a 64-kilodalton subunit of cleavage stimulation factor to mRNA precursors in mouse B cells
Mol. Cell. Biol
.
21
1228
–1238
29
Bruce, S.R., Dingle, R.W., Peterson, M.L.
2003
B-cell and plasma-cell splicing differences: a potential role in regulated immunoglobulin RNA processing
RNA

9
1264
–1273
30
Garneau, D., Revil, T., Fisette, J.F., Chabot, B.
2005
Heterogeneous nuclear ribonucleoprotein F/H proteins modulate the alternative splicing of the apoptotic mediator Bcl-x
J. Biol. Chem
.
280
22641
–22650
31
Tian, B., Hu, J., Zhang, H., Lutz, C.S.
2005
A large-scale analysis of mRNA polyadenylation of human and mouse genes
Nucleic Acids Res
.
33
201
–212
32
Johnson, J.M., Castle, J., Garrett-Engele, P., Kan, Z., Loerch, P.M., Armour, C.D., Santos, R., Schadt, E.E., Stoughton, R., Shoemaker, D.D.
2003
Genome-wide survey of human alternative pre-mRNA splicing with exon junction microarrays
Science

302
2141
–2144
33
Han, K., Yeo, G., An, P., Burge, C.B., Grabowski, P.J.
2005
A combinatorial code for splicing silencing: UAGG and GGGG motifs
PLoS Biol
.
3
e158
34
Huppert, J.L. and Balasubramanian, S.
2005
Prevalence of quadruplexes in the human genome
Nucleic Acids Res
.
33
2908
–2916
35
Todd, A.K., Johnston, M., Neidle, S.
2005
Highly prevalent putative quadruplex sequence motifs in human DNA
Nucleic Acids Res
.
33
2901
–2907
36
Kostadinov, R., Malhotra, N., Viotti, M., Shine, R., D'Antonio, L., Bagga, P.
2006
GRSDB: a database of quadruplex forming G-rich sequences in alternatively processed mammalian pre-mRNA sequences
Nucleic Acids Res
.
34
D119
–D124
37
Zarudnaya, M.I., Kolomiets, I.M., Potyahaylo, A.L., Hovorun, D.M.
2003
Downstream elements of mammalian pre-mRNA polyadenylation signals: primary, secondary and higher-order structures
Nucleic Acids Res
.
31
1375
–1386
38
Kankia, B.I., Barany, G., Musier-Forsyth, K.
2005
Unfolding of DNA quadruplexes induced by HIV-1 nucleocapsid protein
Nucleic Acids Res
.
33
4395
–4403
39
Hazel, P., Huppert, J., Balasubramanian, S., Neidle, S.
2004
J. Am. Chem. Soc
.
126
16405
–16415
40
Petraccone, L., Erra, E., Duro, I., Esposito, V., Randazzo, A., Mayol, L., Mattia, C.A., Barone, G., Giancola, C.
2005
Nucleosides Nucleotides Nucleic Acids

24
757
–760
41
Risitano, A. and Fox, K.R.
2004
Influence of loop size on the stability of intramolecular DNA quadruplexes
Nucleic Acids Res
.
32
2598
–2606
42
D'Antonio, L. and Bagga, P.S.
2004
Computational methods for predicting intramolecular G-quadruplexes in nucleotide sequences Proceedings of the IEEE Conference on Computational Systems Bioinformatics (CSB 2004). pp.
561
–562