![]() | ![]() |
Formats:
|
||||||||||||
Copyright © 2008 RNA Society MicroRNA expression in the adult mouse central nervous system 1Wilhelm Johannsen Center for Functional Genome Research, Department of Cellular and Molecular Medicine, University of Copenhagen, DK-2200 Copenhagen N, Denmark 2Department of Neuroscience and Pharmacology, Panum Institute, University of Copenhagen, DK-2200 Copenhagen N, Denmark 3Institute of Experimental Pathology, ZMBE, University of Münster, D-48149 Münster, Germany 4Santaris Pharma, DK-2970 Hoersholm, Denmark
Reprint requests to: Sakari Kauppinen, Wilhelm Johannsen Center for Functional Genome Research, Department of Cellular and Molecular Medicine, University of Copenhagen, DK-2200 Copenhagen N, Denmark; or Sakari Kauppinen, Santaris Pharma, DK-2970 Hoersholm, Denmark; e-mail:sk/at/imbg.ku.dk; fax: 45-35-327845. Received August 15, 2007; Accepted December 2, 2007. Freely available online through the open access option. This article has been cited by other articles in PMC.Abstract microRNAs are ~22 nucleotide endogenous noncoding RNAs that post-transcriptionally repress expression of protein-coding genes by base-pairing with the 3′-untranslated regions of the target mRNAs. We present here an inventory of miRNA expression profiles from 13 neuroanatomically distinct areas of the adult mouse central nervous system (CNS). Microarray profiling in combination with real-time RT-PCR and LNA (locked nucleic acid)-based in situ hybridization uncovered 44 miRNAs displaying more than threefold enrichment in the spinal cord, cerebellum, medulla oblongata, pons, hypothalamus, hippocampus, neocortex, olfactory bulb, eye, and pituitary gland. These findings suggest that a large number of mouse CNS-expressed miRNAs may be associated with specific functions within these regions. Notably, more than 50% of the identified mouse CNS-enriched miRNAs showed different expression patterns compared to those reported in zebrafish, although the mature miRNA sequences are nearly 100% conserved between the two vertebrate species. The inventory of miRNA profiles in the adult mouse CNS presented here provides an important step toward further elucidation of miRNA function and miRNA-related gene regulatory networks in the mammalian central nervous system. Keywords: microRNA, brain, central nervous system, LNA-ISH, microarray INTRODUCTION MicroRNAs (miRNAs) are small, endogenous noncoding RNA molecules that post-transcriptionally regulate expression of protein-coding genes (Bartel 2004; Kloosterman and Plasterk 2006). To date, 442 murine miRNA sequences have been deposited in the miRBase (Griffiths-Jones et al. 2006), while computational predictions estimate that the vertebrate genomes may contain up to ~1000 miRNA genes (Bentwich et al. 2005; Berezikov et al. 2005). miRNAs are generated from long primary transcripts that are processed in multiple steps to cytoplasmic ~22 nucleotide (nt) mature miRNAs (Bartel 2004; Du and Zamore 2005; Zeng 2006). The mature miRNA is incorporated into the miRNA-induced silencing complex (miRISC), which guides it to target sequences. Most animal miRNAs recognize their target sites located in 3′ UTRs by incomplete base-pairing, resulting in mRNA destabilization or translational repression of the target genes (He and Hannon 2004; Bushati and Cohen 2007). Animal miRNAs have emerged as important players in the control of diverse biological processes (Bartel 2004; Wienholds and Plasterk 2005; Kloosterman and Plasterk 2006). During development, many miRNAs are expressed in neurons or show distinct expression patterns within the developing central nervous system (CNS), implying their importance in brain development and function (Krichevsky et al. 2003; Miska et al. 2004; Sempere et al. 2004; Smirnova et al. 2005; Wienholds et al. 2005). However, functional studies of miRNAs in the vertebrate nervous system are still very limited. Maternal-zygotic dicer mutant zebrafish that lack all mature miRNAs display abnormal brain morphogenesis and neural differentiation (Giraldez et al. 2005). Notably, injection of miR-430 rescues the brain defects in the mutant embryos, inferring a general role in zebrafish brain morphogenesis. In the developing chick neural tube, miR-124a is a component of a regulatory network, which controls the transition between neural progenitors and post-mitotic neurons, by suppressing the anti-neural factor SCP1 (Visvanathan et al. 2007). While miR-124a expression can be detected in E11.5 mouse embryos and it continues to be expressed in neurons of adult mice (Miska et al. 2004; Visvanathan et al. 2007), other miRNAs are temporally expressed during development of the CNS being repressed in the mature CNS (Miska et al. 2004). On the other hand, expression profiling in adult tissues has identified miRNAs enriched in the CNS, suggesting that these miRNAs could play important regulatory roles in mature neurons (Babak et al. 2004; Barad et al. 2004; Miska et al. 2004; Sempere et al. 2004; Thomson et al. 2004). Interestingly, many neuronal miRNAs appear to be localized to actively translating polyribosomes in dendrites, where they may control localized translation of dendrite-specific mRNAs (Kim et al. 2004). This is supported by a study, which showed that miR-134, a brain-specific microRNA, is present in dendrites, where it represses the local synthesis of the protein kinase Limk1 to regulate spine size (Schratt et al. 2006). Stimulation of neurons relieves miR-134-mediated inhibition of Limk1 translation, which, in turn, may contribute to synaptic plasticity (Schratt et al. 2006). Several studies have implicated miRNAs in diseases of the CNS. For example, a mutation in the target site of miR-189 in the human SLITRK1 gene has been shown to be associated with Tourette's syndrome (Abelson et al. 2005), while another study has reported altered miRNA profiles in the prefrontal cortex of patients with schizophrenia and schizoaffective disorder (Perkins et al. 2007). In addition, conditional ablation of Dicer in murine post-mitotic Purkinje cells resulted in progressive loss of miRNAs, cerebellar degeneration, and development of ataxia (Schaefer et al. 2007). Despite the accumulating evidence that miRNAs play important roles in brain development and disorders, our knowledge of miRNA function in the vertebrate nervous system is still very limited. By combining microarray expression profiling with miRNA-specific real-time RT-PCR and LNA-based miRNA in situ detection, we have determined the spatial expression patterns of mouse CNS-expressed miRNAs, which serve as an important basis for detailed studies of individual miRNAs, their target genes, and the miRNA-related regulatory networks in the mammalian central nervous system. RESULTS AND DISCUSSION MicroRNA array profiling of the adult mouse CNS To determine miRNA expression patterns in the adult mouse CNS, 13 different areas of the CNS were dissected from three male balb/c mice: the spinal cord, cerebellum, medulla oblongata, pons, mesencephalon, thalamus, hypothalamus, hippocampus, amygdala, neocortex, olfactory bulb, eye, and pituitary gland. Total RNA samples from these tissues were subsequently pooled, fluorochrome-labeled, and hybridized to spotted miRNA microarrays, comprising LNA-modified probes for all mouse miRNAs in release 7.1 of the miRBase microRNA Registry (Castoldi et al. 2006; Griffiths-Jones et al. 2006). Additionally, RNA from the whole brain of two mice was isolated and analyzed individually. Expression profiling revealed that a large set of miRNAs is expressed in the adult mouse CNS. In agreement with previous reports, several miRNAs, including miR-9, miR-124a, miR-125b, miR-127, miR-128, and members of the let-7 family, were highly enriched in the mouse brain, giving strong hybridization signals on the miRNA arrays (Fig. 1A
Assessment of miRNA expression by miRNA-specific real-time RT-PCR To validate the microarray platform, we assessed the expression of a subset of miRNAs by real-time RT-PCR (Chen et al. 2005), using the same RNA samples that were applied to the microarrays. These included five differentially expressed miRNAs: miR-200a (olfactory bulb), miR-200c (olfactory bulb), miR-205 (eye), miR-195 (cerebellum), and miR-124a (absent from pituitary), as well as three miRNAs exhibiting a more uniform expression (let-7a, let-7d, and miR-29c). We found strong correlation between our microarray profiling and real-time RT-PCR data (R
2 = 0.63; R
2 = 0.93 when removing two outliers) (data not shown). Next, we analyzed the expression patterns of seven miRNAs by real-time RT-PCR assays in 13 dissected CNS regions from three additional mice (Fig. 3A–G
In situ detection of miRNA accumulation in the mouse CNS The spatial expression patterns of two miRNAs identified as differentially expressed by microarray profiling were determined by in situ hybridization (ISH) using LNA probes (Fig. 3H,I Coordinated expression of miRNAs and their host genes Many miRNAs located within protein-coding and non-protein-coding genes are transcriptionally linked to the expression of their host genes (Rodriguez et al. 2004). In order to investigate the coordinated expression of the differentially expressed miRNAs identified in this study with their predicted host transcripts, we compiled the mRNA expression data of the relevant protein-coding genes, which are summarized in Table 1 (GNF SymAtlas version 1.2.4) (Su et al. 2004). A large group of the CNS region-specific miRNAs that reside within other genes shows highest expression levels in the same regions as their host genes, implying that they are cotranscribed. For example, the hippocampus-enriched miR-218-1 and miR-218-2 genes are located within Slit2 and Slit3. Accordingly, the Slit3 gene displays highest expression in the hippocampus. Another example is the pituitary-specific miR-152, which at the genomic level is localized within the Copz2 gene, which also shows most prevalent expression in the pituitary. Furthermore, our results infer miR-204 as highly enriched in the eye, which is in good agreement with a previous study demonstrating coexpression of miR-204 and its host gene Trpm3 in adult mouse eye (Karali et al. 2007). Finally, miR-10a appears to be more prevalent in the spinal cord compared to other regions. This is consistent with miR-10a being located within Hoxd4 (ENSMUST00000047904), which also shows highest expression levels in the spinal cord and medulla oblongata. However, it is important to note that failure of identifying coexpression of miRNA and the host gene does not exclude the possibility that they share the same set of transcriptional control elements. Differences in turnover or processing of a miRNA and its host gene could result in highly different expression levels within the same tissue. miRNAs that are closely linked at the genomic level often exhibit coordinated expression between different tissues, indicating that they share common cis-regulatory elements or are derived from polycistronic precursors (Sempere et al. 2004; Baskerville and Bartel 2005). In the present study, we found high correlation of expression of the miRNA cluster hosted by the protein-coding gene Ttll10: miR-429|miR-200a|miR-200b (R = 0.81–0.89) and the following, independently transcribed miRNA clusters: (1) miR-221|miR-222 (R = 0.89), (2) miR-96|miR-183 (R = 0.92), (3) miR-200c|miR-141 (R = 0.89), (4) miR-195|miR-497 (R = 0.86), and (5) miR-99a|let-7c (R = 0.80). Expression of miRNAs and their predicted targets in the mouse CNS A major challenge in understanding the biology of microRNAs is to identify their target genes. While plant miRNAs are generally perfectly complementary to their target mRNAs, most animal miRNAs pair to the 3′ UTRs of their targets by incomplete base-pairing, in which nucleotides 2–7 of the mature miRNA sequence, termed the seed region, appear to be critical for target site recognition (Lewis et al. 2005). Previous computational analyses of microarray data have shown that predicted mRNA targets of several highly tissue-specific miRNAs are expressed at significantly lower levels in the same tissues compared to tissues where such miRNAs are not expressed (Farh et al. 2005; Sood et al. 2006). For example, miR-1 is highly prevalent in the heart and skeletal muscle, whereas the predicted targets of miR-1 are expressed at significantly lower levels in heart and skeletal muscle compared to other tissues (Sood et al. 2006). This can be explained by miRNA-mediated destabilization of target mRNA levels, which lends experimental support from studies reporting degradation of large numbers of target mRNAs upon transfection of exogenous miRNA into cells or de-repression of targets upon antagonizing specific miRNAs by antagomirs in vivo (Krutzfeldt et al. 2005; Lim et al. 2005). On the basis of depletion of 7-mer seed sites in the 3′-UTRs of mammalian mRNAs, Farh et al. (2005) predicted the expression signatures of 73 miRNA families conserved among the four sequenced mammals and zebrafish in 61 tissues. In this study, we were able to experimentally confirm the predicted expression patterns for many of the aforementioned miRNAs (Table 1). For example, miR-375 and miR-7 are predicted to be expressed in the pituitary and, indeed, our results show highest accumulation of both miR-7 and miR-375 in the pituitary. Furthermore, miR-96, miR-200a, miR-200b, miR-141, and miR-183 all are predicted to be expressed in the olfactory epithelium, which is consistent with our results showing enrichment of these miRNAs in the olfactory bulb. We also find miR-10a and miR-10b to be enriched in both medulla oblongata and spinal cord, which is consistent with their predicted accumulation in the lower spinal cord (Farh et al. 2005). Notably, we also find enriched miRNAs in CNS regions in which they were not predicted to be expressed (Table 1). Our findings that miR-34a is more highly expressed in the medulla oblongata, pons, and spinal cord compared to other regions of the CNS, and that miR-204 and miR-205 are prevalently expressed in the eye, suggest that these miRNAs might be coexpressed with their targets. Regional coexpression of miRNAs and their targets has been reported. Coexpression of miR-200b and one of its targets Zfhx1b, as well as miR-189 and its target Slitrk1, is observed in several areas of the adult mouse brain (Abelson et al. 2005; Christoffersen et al. 2007). Additionally, luciferase reporter assays have identified myotrophin (Mtpn) as a target of miR-124 regulation (Krek et al. 2005), both of which are highly expressed in neurons throughout the brain (Fujigasaki et al. 1996). Hence, it is tempting to speculate that miRNAs expressed in the same tissues as their target genes might function by fine-tuning their expression rather than by completely suppressing them. Comparison of miRNA expression between zebrafish and mouse Highly divergent expression patterns of conserved miRNAs in zebrafish, medaka, chicken, and mouse have previously been reported (Ason et al. 2006). For example, while miR-125b is ubiquitously expressed in the brain and spinal cord of zebrafish, its expression is confined to the mid-hindbrain boundary in the mouse (Ason et al. 2006). Comparison of the expression profiles of the 44 differentially expressed mouse CNS miRNAs identified in this study with those reported by Wienholds et al. (2005) revealed conserved expression for 15 out of 36 miRNAs between zebrafish and mouse (Table 1). Examples of miRNAs with conserved expression include miR-200b and miR-375 and miR-204, which are enriched in the olfactory bulb, pituitary, and eye, respectively, as well as miR-96, which is enriched in the olfactory bulb and eye. In contrast, the expression patterns of 21 mouse CNS-enriched miRNAs appeared to be divergent from those of zebrafish, although the mature miRNA sequences are nearly 100% conserved between the two vertebrates, while all except four miRNAs show 100% conservation in their seed regions (Table 1). Clear examples of divergent expression patterns between zebrafish and mouse CNS are miR-31, which is enriched in the eye in mice, whereas it is ubiquitously expressed in zebrafish; and miR-195, which in mice is enriched in the cerebellum, whereas in zebrafish it displays widespread expression. Additionally, miR-142-3p expression is enriched in the medulla oblongata and spinal cord in mice but confined to the thymus and blood cells in zebrafish. The mammalian CNS is highly complex with a broad, fine-tuned network of molecular interactions, in which processes such as learning and memory, neuronal repair, and regeneration are dependent on highly orchestrated gene expression programs. miRNAs have emerged as important post-transcriptional regulators of developmental and physiological processes, including neuronal differentiation and brain development and function. The previously reported differences in vertebrate miRNA expression patterns among four vertebrate species (Ason et al. 2006) along with our findings here may reflect differences in species physiology, including the complex cell-type compositions of the vertebrate nervous systems. Recent reports suggest that miRNAs can regulate dendritic spine size (Schratt et al. 2006) and neuronal morphogenesis (Vo et al. 2005), whereas massive parallel sequencing of small RNA libraries from human and chimpanzee brain has revealed highly complex miRNA repertoires in the primate brain (Berezikov et al. 2006). It is therefore tempting to speculate that miRNAs could play important roles in the complex gene-regulatory circuits of the mammalian CNS and may provide an important contribution to evolution of biological complexity. In conclusion, the inventory of miRNA profiles in the adult mouse CNS presented here provides an important step toward further elucidation of miRNA function and miRNA-related gene regulatory networks in the mammalian central nervous system. MATERIALS AND METHODS Mouse tissues Brain regions from adult male Balb/c mice were dissected with RNaseZap (Ambion) treated tools and immediately transferred to RNAlater medium (Ambion). RNA was isolated using Trizol reagent (Invitrogen) as described by the manufacturer, except that 85% ethanol instead of 75% ethanol was used to wash the RNA pellet. Additionally, 10–15 μg of glycogen (Ambion) was added as carrier prior to precipitation. RNA integrity was assessed on 2% agarose gels stained with ethidium bromide and quantified using a Ribogreen RNA quantification kit (Invitrogen) and a fluorometer (Thermo Scientific). Microarray printing, labeling, and hybridization LNA-modified oligonucleotide probes for all mouse microRNAs annotated in miRBase version 7.1 were obtained from Exiqon (miRCURY version 7.1; Exiqon). Probes were diluted to a final concentration of 10 μM in printing buffer (150 mM sodium phosphate at pH 8.5) and printed onto Codelink slides (GE Healthcare) using a MicroGrid TAS II arrayer (Biorobotics). Spotted slides were post-processed according to the manufacturer's recommendations. Total RNA (2–4 μg) was 3′-end-labeled using T4 RNA ligase and a Cy3-labeled RNA linker (Cole et al. 2004; Wienholds et al. 2005) by the following procedure: RNA in 4.5 μL of water was combined with 0.8 μL of T4 RNA ligase buffer (10×) (Ambion), 1.1 μL of polyethyleneglycol (50% [w/v]), 0.8 μL of RNA-linker (250 μM; DNA Technology), and 0.8 μL of T4 RNA ligase (Ambion). The reaction was incubated for 2 h at 30°C, and terminated by incubation for 3 min at 80°C. Labeled RNA (8 μL) was combined with 6 μL of 20× SSC (Ambion), 1.5 μL of herring sperm DNA (10 mg/mL; Roche), 11.4 μL of formamide (Sigma), 0.6 μL of 5% SDS (Ambion), and 2.5 μL of DEPC-treated water. Samples were denatured for 1–2 min at 80°C and hybridized to the microarray for 16–20 h at 65°C under a lifterslip (Erie Scientific). Post-hybridization washes were in 4× SSC at 60°C to remove the coverslip, followed by three times in 2× SSC, 0.025% SDS for 5 min each, three times in 0.8× SSC for 2 min each, and two times in 0.4× SSC for 3 min each. Microarray data analysis Microarray slides were scanned using an ArrayWorx Biochip Reader (Applied Precision). Scanning images were analyzed using GridGrinder (http://gridgrinder.sourceforge.net/). Background-subtracted spot intensities were normalized using variance stabilization normalization (Huber et al. 2002). Significance analysis of microarrays (SAM) was used to identify miRNAs differentially expressed between samples (FDR = 0) (Tusher et al. 2001). Brain-area-enriched miRNAs were identified by one-sample t-tests using pooled variances (P < 0.05). P-values were adjusted for multiple comparisons using the Bonferroni procedure. Clustering and visualization of expression data were done with MultiExperimentViewer (www.tm4.org) (Saeed et al. 2003). Real-time RT-PCR Real-time RT-PCR analyses were carried out using TaqMan MicroRNA Assays (Applied Biosystems) according to the manual. Relative expression was calculated using the ΔΔCT method (Livak and Schmittgen 2001) and normalized to the expression of snoRNA202 (Applied Biosystems). Detection of spatial miRNA accumulation by in situ hybridization In situ hybridizations were performed in 10-μm cryosections from adult mouse brain. Sections were fixed in 4% paraformaldehyde and acetylated in acetic anhydride/triethanolamine, followed by washes in PBS. Sections were then pre-hybridized in hybridization solution (50% formamide, 5× SSC, 0.5 mg/mL yeast tRNA, 1× Denhardt's solution) at 25°C below the predicted T
m value of the LNA probe for 30 min. Probes (3 pmol) (LNA miRCURY probe; Exiqon) were DIG-labeled (DIG Oligonucleotide 3′ Tailing Kit; Roche Applied Sciences) and hybridized to the sections for 1 h at the same temperature as pre-hybridization. After post-hybridization washes in 0.1× SSC at 55°C, the in situ hybridization signals were detected using the tyramide signal amplification system (PerkinElmer) according to the manufacturer's instructions. Slides were mounted in Prolong Gold containing DAPI (Invitrogen) and analyzed with an Olympus MVX10 microscope equipped with a CCD camera and Olympus CellF software. ACKNOWLEDGMENTS This study is supported by grants from the Lundbeck Foundation, the Danish National Advanced Technology Foundation, the Læge Sofus Carl Emil Friis and hustru Olga Friis' Legat, the Nationales Genomforschungsnetz (NGFN, 0313358A), and the European Commission as part of the RIBOREG EU FP6 project (LSHG-CT-2003503022). Wilhelm Johannsen Centre for Functional Genome Research is established by the Danish National Research Foundation. Footnotes Article published online ahead of print. Article and publication date are at http://www.rnajournal.org/cgi/doi/10.1261/rna.783108. REFERENCES
|
PubMed related articles
Your browsing activity is empty. Activity recording is turned off. |
|||||||||||
Cell. 2004 Jan 23; 116(2):281-97.
[Cell. 2004]Dev Cell. 2006 Oct; 11(4):441-50.
[Dev Cell. 2006]Nucleic Acids Res. 2006 Jan 1; 34(Database issue):D140-4.
[Nucleic Acids Res. 2006]Nat Genet. 2005 Jul; 37(7):766-70.
[Nat Genet. 2005]Cell. 2005 Jan 14; 120(1):21-4.
[Cell. 2005]Cell. 2004 Jan 23; 116(2):281-97.
[Cell. 2004]FEBS Lett. 2005 Oct 31; 579(26):5911-22.
[FEBS Lett. 2005]Dev Cell. 2006 Oct; 11(4):441-50.
[Dev Cell. 2006]RNA. 2003 Oct; 9(10):1274-81.
[RNA. 2003]Genome Biol. 2004; 5(9):R68.
[Genome Biol. 2004]Science. 2005 Oct 14; 310(5746):317-20.
[Science. 2005]Genome Biol. 2007; 8(2):R27.
[Genome Biol. 2007]J Exp Med. 2007 Jul 9; 204(7):1553-8.
[J Exp Med. 2007]RNA. 2006 May; 12(5):913-20.
[RNA. 2006]Nucleic Acids Res. 2006 Jan 1; 34(Database issue):D140-4.
[Nucleic Acids Res. 2006]RNA. 2004 Nov; 10(11):1813-9.
[RNA. 2004]Genome Res. 2004 Dec; 14(12):2486-94.
[Genome Res. 2004]Genome Biol. 2004; 5(9):R68.
[Genome Biol. 2004]Genome Biol. 2004; 5(3):R13.
[Genome Biol. 2004]RNA. 2005 Sep; 11(9):1461-70.
[RNA. 2005]Nucleic Acids Res. 2005 Nov 27; 33(20):e179.
[Nucleic Acids Res. 2005]Science. 2005 Jul 8; 309(5732):310-1.
[Science. 2005]Nat Methods. 2006 Jan; 3(1):27-9.
[Nat Methods. 2006]Nat Protoc. 2007; 2(6):1508-14.
[Nat Protoc. 2007]Nat Protoc. 2007; 2(10):2520-8.
[Nat Protoc. 2007]Genome Res. 2004 Oct; 14(10A):1902-10.
[Genome Res. 2004]Proc Natl Acad Sci U S A. 2004 Apr 20; 101(16):6062-7.
[Proc Natl Acad Sci U S A. 2004]Invest Ophthalmol Vis Sci. 2007 Feb; 48(2):509-15.
[Invest Ophthalmol Vis Sci. 2007]Genome Biol. 2004; 5(3):R13.
[Genome Biol. 2004]RNA. 2005 Mar; 11(3):241-7.
[RNA. 2005]Cell. 2005 Jan 14; 120(1):15-20.
[Cell. 2005]Science. 2005 Dec 16; 310(5755):1817-21.
[Science. 2005]Proc Natl Acad Sci U S A. 2006 Feb 21; 103(8):2746-51.
[Proc Natl Acad Sci U S A. 2006]Nature. 2005 Dec 1; 438(7068):685-9.
[Nature. 2005]Nature. 2005 Feb 17; 433(7027):769-73.
[Nature. 2005]Proc Natl Acad Sci U S A. 2006 Sep 26; 103(39):14385-9.
[Proc Natl Acad Sci U S A. 2006]Science. 2005 Jul 8; 309(5732):310-1.
[Science. 2005]Proc Natl Acad Sci U S A. 2006 Sep 26; 103(39):14385-9.
[Proc Natl Acad Sci U S A. 2006]Nature. 2006 Jan 19; 439(7074):283-9.
[Nature. 2006]Proc Natl Acad Sci U S A. 2005 Nov 8; 102(45):16426-31.
[Proc Natl Acad Sci U S A. 2005]Nat Genet. 2006 Dec; 38(12):1375-7.
[Nat Genet. 2006]Nucleic Acids Res. 2004; 32(11):e86.
[Nucleic Acids Res. 2004]Science. 2005 Jul 8; 309(5732):310-1.
[Science. 2005]Bioinformatics. 2002; 18 Suppl 1():S96-104.
[Bioinformatics. 2002]Proc Natl Acad Sci U S A. 2001 Apr 24; 98(9):5116-21.
[Proc Natl Acad Sci U S A. 2001]Biotechniques. 2003 Feb; 34(2):374-8.
[Biotechniques. 2003]Methods. 2001 Dec; 25(4):402-8.
[Methods. 2001]