This Article Abstract Full Text (PDF) Alert me when this article is cited Alert me if a correction is posted Services Similar articles in this journal Similar articles in PubMed Alert me to new issues of the journal Download to citation manager Citing Articles Citing Articles via HighWire Citing Articles via Google Scholar Google Scholar Articles by Lal, A. Articles by Riggins, G. J. Search for Related Content PubMed PubMed Citation Articles by Lal, A. Articles by Riggins, G. J.

A Public Database for Gene Expression in Human Cancers¹

Department of Pathology, Duke University Medical Center, Durham, North Carolina 27710 [A. L., R. E. M., G. J. R.]; National Center for Biotechnology Information, National Library of Medicine, NIH, Bethesda, Maryland 20894 [A. E. L., S. F. A.]; The Johns Hopkins Oncology Center [V. V., K. W. K.], The Howard Hughes Medical Institute [L. Z., B. V.], and Department of Pathology [P. J. M.], Johns Hopkins University School of Medicine, Baltimore, Maryland 21231; Washington University Genome Sequencing Center, St. Louis, Missouri 63108 [M. A. M.]; The I.M.A.G.E. Consortium, Biology and Biotechnology Research Program, Lawrence Livermore National Laboratory, Livermore, California 94550 [C. P.]; Laboratory of Biological Chemistry, Gerontology Research Center, National Institute on Aging, Baltimore, Maryland 21224 [P. J. M.]; Department of Adult Oncology, Dana-Farber Cancer Institute, Boston, Massachusetts 02115 [K. P.]; Department of Pathology, Columbia University, New York, New York 10032 [N. P.]; and Cancer Genome Anatomy Project, Office of the Director, National Cancer Institute, Bethesda, Maryland 20892 [R. L. S.]

	ABSTRACT

Top ABSTRACT Introduction Materials and Methods Results and Discussion REFERENCES

 
A public database, SAGEmap, was created as a component of the Cancer Genome Anatomy Project to provide a central location for depositing, retrieving, and analyzing human gene expression data. This database uses serial analysis of gene expression to quantify transcript levels in both malignant and normal human tissues. By accessing SAGEmap (http://www.ncbi.nlm.nih.gov/SAGE) the user can compare transcript populations between any of the posted libraries. As an initial demonstration of the database’s utility, gene expression in human glioblastomas was compared with that of normal brain white matter. Of the 47,174 unique transcripts expressed in these two tissues, 471 (1.0%) were differentially expressed by more than 5-fold (P < 0.001). Classification of these genes revealed functions consistent with the biological properties of glioblastomas, in particular: angiogenesis, transcription, and cell cycle related genes.

	Introduction

Top ABSTRACT Introduction Materials and Methods Results and Discussion REFERENCES

 
The advent of new technologies has made analysis of overall gene expression patterns a practical reality (1, 2, 3) . A major goal of CGAP⁴ has been the identification of genes that undergo alteration in expression during malignant transformation, in part by capitalizing on these recent technical advances.⁵ However, the means to efficiently generate, publicly distribute, and analyze comprehensive gene expression data from multiple laboratories did not exist previously. SAGE was chosen to create a comprehensive quantitative expression database for CGAP, because this assay provides absolute transcript numbers in a digital format (3 , 4) and can be easily adapted to provide statistical comparisons of data from multiple laboratories. The basis of SAGE is to count expressed transcripts by sequencing a 14-bp ‘tag’ of the gene, which is normally sufficient for gene identification. The four bases of the 3'-most NlaIII restriction site plus the following ten variable bases define the transcript tag. Past SAGE experiments have revealed that a small, but likely important, fraction of a tissue’s expressed genes has sustained a substantial change in expression after malignant transformation (5 , 6) . Unfortunately, the large data sets from these studies are not easily accessible for most researchers, and the time and cost of generating new tumor specific data are generally prohibitive. The goal of this project was to create the means to disseminate to the research community, without restrictions, comprehensive gene expression data for the study of most common human cancers. To initiate this first public quantitative gene expression database, we cloned and sequenced SAGE tags from 25 different normal and tumor tissues and created a system for rapid transfer of the data to the SAGEmap web site. Custom tools were created for this web site to analyze the expression of a single gene in multiple tissues or to compare the full expression pattern between tissues. Furthermore, we demonstrate one use of this database by reporting differentially expressed genes found by comparing glioblastoma with normal brain tissue. This is one of many comparisons that may be performed on SAGEmap.

	Materials and Methods

Top ABSTRACT Introduction Materials and Methods Results and Discussion REFERENCES

 
Tumor and Normal Samples.
The following samples were used to construct the first 25 SAGEmap libraries: two colorectal carcinoma primary tumors, four colorectal cancer cell lines, two normal colon epithelium samples, four ovarian carcinoma cell lines, one normal ovary surface epithelium, two ovarian tumors, one prostate cell line, one GBM primary tumor, one pool of five GBM primary tumors, one fibrillary astrocytoma, one glioblastoma cell line, two medulloblastoma cell lines, one normal human astrocyte culture, and two normal brain cortical white matter samples. One normal brain white matter library was made from a rapid autopsy sample with a postmortem delay of 3 h and 15 min and confirmed normal histology (provided by the Bryan Alzheimer Disease Research Center, Duke University). The other normal brain library was constructed from normal surrounding brain removed during the surgical approach to remove a seizure focus or a glioblastoma. For each glioblastoma sample a neuropathologist (R. E. M.) confirmed the appearance of a GBM (Grade IV astrocytoma), without necrosis, by examining adjacent stained sections from the sample margins. Immunohistochemistry and slide histology were performed on fixed tissue sections embedded in paraffin. HAM56, a mouse monoclonal antibody (DAKO Corp., Carpinteria, CA), was used to identify macrophage specific markers in primary GBM tissue sections.

SAGE.
Total RNA for use in the SAGE protocol was prepared from tissue or cell lines by ultracentrifugation over a cesium chloride gradient followed by mRNA selection by oligo(dT) cellulose (Life Technologies, Gaithersburg, MD). The SAGE libraries were constructed as described previously (3 , 7) . For the brain libraries, 24 colonies were screened from each library before large-scale sequencing to ensure that they had an insert containing SAGE ditags. Approximately 2500 bacterial colonies were randomly picked by a robot and arrayed in 384-well plates. Plasmids from each colony were purified and sequenced as described previously (8) . Transcript tags were extracted from the sequence files, using the SAGE software, v3.03. Tags matching linker sequence (approximately 4%) and duplicate ditags were excluded before analysis or posting on the web site. To estimate the total number of expressed genes in each library, the unique tags were matched to a list of presumed possible ‘real’ tags.⁶ This list compromises only those tags that are unlikely to occur by random sequencing errors, in part, due to the number of times they were observed in a pool of 3.5 million transcripts. Statistical comparisons of transcript numbers and P values presented here were derived from the Monte-Carlo analysis in the SAGE software (5) . SAGE tags from colon libraries were a combination of newly generated tags for this project or from a previous study (5) .

Northern Blotting and RT-PCR.
For northern analysis, total RNA was isolated by ultracentrifugation over a cesium chloride gradient from five normal brain samples, seven glioblastoma primary tumors, three cell lines, and six xenografts. Radioactive labeled PCR probes were made from the following genes (with GenBank accession number in parentheses):

(a) -glucosidase (X87237);

(b) p1-cdc46 (X74795);

(c) cartilage-39 glycoprotein (M80927);

(d) nmb (X76534);

(e) nicotinamide N-methyltransferase (U08021);

(f) SPARC (J03040);

(g) insulin-like growth factor binding protein 2 (M35410);

(h) CAPL (M80563);

(i) CSRP2 (U57646);

(j) Eph-family protein (D83492);

(k) SH3GL2 (X99657);

(l) hCDCrel-1 (U74628); and

(m) EST (AI214960).

To confirm expression changes by PCR, total RNA was converted to random-primed cDNA using reverse-transcriptase for a panel of three normal brain samples, six glioblastoma primary tumors, two cell lines, and two xenografts. PCR of each cDNA was performed for 20, 25, 28, and 32 cycles using primers specific for -actin, lipocortin (GenBank accession X05908), three ESTs (GenBank accessions AI302970, AI022985, and AA903288), and dynamin (GenBank accession L07807). The PCR reactions were normalized to the -actin levels, and the products were compared on gel electrophoresis by band intensity to confirm large changes in gene expression between tumor and normal.

SAGEmap Informatics.
Tag-to-gene mapping was accomplished by first orienting GenBank sequences using polyadenylation signal and tail and placing the sequences into five "confidence" classes. In each of these confidence classes, tags were extracted from the ten-base tag directly 3'-adjacent to the 3'-most NlaIII site, and then linked to a UniGene cluster identifier, based on the sequence’s UniGene cluster assignment. This full mapping was used to provide the tag-to-gene and gene-to-tag information on the web site. In addition, a ‘reliable’ tag-to-gene mapping was designed to reduce the effects of sequencing errors in EST GenBank entries. To correct for these errors, tag-gene pairings were first ordered by the frequency that a tag was observed in a particular cluster. Assuming the errors occur randomly and a sequencing error rate of 1% per base, 10 bases have roughly a 10% chance [1-(0.99)¹⁰] of having one or more errors. Therefore, the lowest 10% of the rank-ordered tag-gene pairings were eliminated. It was this mapping that is used to make tag-gene assignments for SAGE tag frequency on the web site. The above approach was written into an automatic algorithm, which is used to update SAGEmap weekly.

To avoid the random simulations required to assess statistical significance by the SAGE software (5) , which would be impractical for an interactive web site, SAGEmap uses a previously described Bayesian approach (9) . We have extended this method so that it can deal with unequal aggregate numbers of tags A and B from the two libraries being compared. For an mRNA species with concentrations y and z in these libraries, we study the quantity x = y ÷ (y + z), which we assume has a prior probability density f(x) over the interval zero to one. This function captures knowledge about the distribution of relative mRNA concentrations for the complete set of mRNA species. It is reasonable to assume that f(x) will peak at, and be symmetric about, 0.5 (5 , 10) , and, for simplicity, f(x) may be taken proportional to x^c(1 - x)^c. Setting c = 1 has been proposed (9) , but various data (5 , 10) suggest a greater concentration of x near 0.5, implying c = 3 is a more appropriate value. If a and b copies of a tag corresponding to a specific mRNA are sequenced from the two libraries, the posterior probability density for x (i.e., the density given these data) is proportional to g(x), given by the following equation (derivation omitted).

One may calculate from g(x) the posterior probability that the concentration y exceeds z by any given factor. Specifically, y exceeds z by at least the factor F when x L, where L = F ÷ (F + 1). The desired probability is simply the proportion of g(x), viewed as a density, that falls greater than L. The posterior probability that this is the case is given by the following equation.

	Results and Discussion

Top ABSTRACT Introduction Materials and Methods Results and Discussion REFERENCES

 
More than 1,000,000 transcript tags from 25 different tumor or normal tissue SAGE libraries were cloned, sequenced, and deposited in the SAGEmap database using the project organization shown in Fig. 1 . Six of these libraries were designed to study glioblastoma pathogenesis. Some of the possible transcript number comparisons from these libraries are shown in Table 1 . From these comparisons we sought one that might provide insight into the pattern of expression necessary for glioblastoma maintenance or formation. Glioblastoma is a deadly, highly malignant form of brain cancer thought to arise mostly from astrocytes present in cortical white matter (11) . Therefore, we chose for further analysis the expression comparison between surgically removed glioblastomas and normal brain white matter. Tag counts generated by sequencing the libraries were used to determine transcript expression levels and to find statistically significant differences between the expressed genes. This comparison yielded the expected distribution (5) of the 282,760 SAGE tags (Fig. 2) and showed that 1.0% (471) of the unique expressed genes had more than a 5-fold expression change between the tissues, and a P < 0.001.

View larger version (13K): [in this window] [in a new window]   Fig. 1. Schematic of data flow to and from the SAGEmap web site. SAGE libraries from various tumors, normal tissue, and cell lines are constructed at the various laboratories, and the bacterial clones are plated and shipped for arraying and sequencing. From the raw sequencing data, the ten bp of unique SAGE tag sequence is extracted and used as a gene identifier. The National Center for Biotechnology Information manages the informatics and links a UniGene cluster to each tag. Web site users can define groups of libraries and statistical parameters for on-line comparisons. It is also possible to infer gene expression levels by determining a gene’s SAGE tag and looking at its frequency in the different libraries, using a virtual Northern function. Mapping information is provided as a resource to SAGE users to enhance the ability to locate the correct gene from a SAGE tag. The data are fully downloadable, for local analysis.

View larger version (65K): [in this window] [in a new window]   Fig. 2. Distribution of the SAGE tags from surgically resected glioblastomas and normal white-matter libraries. The number of times each unique SAGE tag was observed was plotted on a logarithmic scale, using a total of 107,349 tags from surgically resected glioblastomas (y-axis) versus 123,880 tags for primary brain white matter (x-axis). Tags with no expression in one of the two groups were set to a value of one; and the tag populations were normalized. The line with slope of one through the center predicts equal gene expression in the two tissues. The P that an observed difference was real was calculated for each gene. Those genes with a significant P less than 0.001 (‘+’ symbol in black) are furthest from the line, whereas more than 99% of the genes fell closer with a P between 0.001 and 0.01 () or P greater than 0.01 ().

Proper tag-to-gene mapping was obvious when the tag identified an accurate full-length sequence, such as most of the named GenBank entries. However, sequencing errors in ESTs can create ambiguity in the mapping process. To enhance tag-to-gene mapping for ESTs, an algorithm based on UniGene⁷ was constructed to ascertain mapping reliability. UniGene has sequence similarity clusters built from overlapping sequences and may have different tag sequences within the same cluster. However, SAGEmap takes into account the number of times a particular tag sequence is observed within each cluster, the orientation of the sequence, and the presence of a polyadenylation signal and tail. Thus, the likelihood that a particular tag is real and is not generated by sequencing errors is assessed and used for proper mapping.

We also constructed a tool for on-line statistical comparisons of tag populations for SAGEmap (Fig. 3A) . This function allows comparisons between any possible combination of libraries and provides a confidence level for each differentially expressed tag. Desired fold differences can be specified to select the desired tags and libraries for comparison to suit the particular needs of an investigator. A ‘virtual northern’ function is available that calculates the fractional representation of a particular gene in all of the posted libraries (Fig. 3B) . In addition, tag frequencies and mapping information are downloadable for local analysis.

View larger version (64K): [in this window] [in a new window]   Fig. 3. Screen images from the SAGEmap web site. Users can select libraries or groups of libraries for comparison and view lists of differentially expressed tags (A). In this case, only those genes with greater than 5-fold difference were included in the search parameters. Over- or underexpressed genes are color-coded in green or red, the number of tags in each group listed, a measure provided of the variation of the tag number within a group (CoV), a level of statistical confidence that the difference is not due to random chance, and a hyperlink to the relevant tag containing UniGene cluster is provided. SAGEmap users can paste into the virtual Northern tool a sequence of interest (containing the 3' end of the gene) and the web site will extract the likely tag and show its expression level in all of the available SAGE libraries. Tags produced from library comparisons are also linked to this function. The result of creating a virutual Northern for VEGF, from a known sequence pasted from GenBank, is shown in B.

SAGEmap is the first completely public database for quantitative gene expression comparisons and provides a central repository for SAGE data. Most major tumor types and normal tissues are planned for representation on this database as well as comparisons designed to provide insight into the major genetic pathways involved in malignant transformation (7 , 16, 17, 18) . Another practical advantage of this data is that they provide candidate tumor-specific genes for immune-based cancer therapy (19 , 20) . SAGEmap now makes some large-scale expression data and comparisons quickly accessible to any researcher with Internet access. These data demonstrate one aspect of the utility of a gene expression database that provides a prototype for future reporting and dissemination of quantitative gene expression data. It is hoped that the virtual experiments possible using this database will accelerate certain aspects of cancer research.

	ACKNOWLEDGMENTS

 
We thank D. Bigner, S. Bigner, H. Friedman, and the Duke Neuro-Oncology Program members for assistance in acquiring samples. G. J. R. is a James S. McDonnell Foundation Scholar and a Novartis Faculty Scholar. Work by C. P. was supported by the United States Department of Energy contract W-7405-Eng-48 to Lawrence Livermore National Laboratory.

	FOOTNOTES

 
The costs of publication of this article were defrayed in part by the payment of page charges. This article must therefore be hereby marked advertisement in accordance with 18 U.S.C. Section 1734 solely to indicate this fact.

¹ Funding for this work was provided by Cancer Genome Anatomy Project (contract S98-146A).

² These authors contributed equally to this work.

³ To whom requests for reprints should be addressed, at Duke University Medical Center, Box 3156, Durham, NC 27710. Phone: (919) 684-5343; Fax: (919) 681-2796; E-mail: greg.riggins{at}duke.edu

⁴ The abbreviations used are: CGAP, Cancer Genome Anatomy Project; SAGE, Serial Analysis of Gene Expression; EST, expressed sequence tag; GBM, glioblastoma multiforme; RT-PCR, reverse-transcriptase PCR; VEGF, vascular endothelial growth factor.

⁵ See http://www.ncbi.nlm.nih.gov/ncicgap.

⁶ V. E. Velculescu, S. L. Madden, L. Zhang, A. E. Lash, J. Yu, C. Rago, A. Lal, C. J. Wang, G. A. Beaudry, K. M. Ciriello, B. P. Cook, M. R. Dufault, A. T. Ferguson, Y. Gao, T. C. He, H. Hermeking, S. K. Hiraldo, P. M. Hwang, M. A. Lopez, H. F. Luderer, B. Mathews, J. M. Petroziello, K. Polyak, L. Zawel, W. Zhang, X. Zhang, W. Zhou, F. G. Haluska, J. Jen, S. Sukumar, G. M. Landes, G. J. Riggins, B. Vogelstein, and K. W. Kinzler. Analysis of human transcriptones, submitted for publication.

⁷ See http://www.ncbi.nlm.nih.gov/UniGene.

Received 7/20/99. Accepted 9/21/99.

	REFERENCES

Top ABSTRACT Introduction Materials and Methods Results and Discussion REFERENCES

 

1. Lockhart D. J., Dong H., Byrne M. C., Follettie M. T., Gallo M. V., Chee M. S., Mittmann M., Wang C., Kobayashi M., Horton H., Brown E. L. Expression monitoring by hybridization to high-density oligonucleotide arrays. Nat. Biotechnol., 14: 1675-1680, 1996.[Medline]
2. Schena M., Shalon D., Davis R. W., Brown P. O. Quantitative monitoring of gene expression patterns with a complementary DNA microarray. Science (Washington DC), 270: 467-470, 1995.[Abstract/Free Full Text]
3. Velculescu V. E., Zhang L., Vogelstein B., Kinzler K. W. Serial analysis of gene expression. Science (Washington DC), 270: 484-487, 1995.[Abstract/Free Full Text]
4. Velculescu V. E., Zhang L., Zhou W., Vogelstein J., Basrai M. A., Bassett D. E., Jr, Hieter P., Vogelstein B., Kinzler K. W. Characterization of the yeast transcriptome. Cell, 88: 243-251, 1997.[Medline]
5. Zhang L., Zhou W., Velculescu V. E., Kern S. E., Hruban R. H., Hamilton S. R., Vogelstein B., Kinzler K. W. Gene expression profiles in normal and cancer cells. Science (Washington DC), 276: 1268-1272, 1997.[Abstract/Free Full Text]
6. Hibi K., Liu Q., Beaudry G. A., Madden S. L., Westra W. H., Wehage S. L., Yang S. C., Heitmiller R. F., Bertelsen A. H., Sidransky D., Jen J. Serial analysis of gene expression in non-small cell lung cancer. Cancer Res., 58: 5690-5694, 1998.[Abstract/Free Full Text]
7. Hermeking H., Lengauer C., Polyak K., He T. C., Zhang L., Thiagalingam S., Kinzler K. W., Vogelstein B. 14–3-3 is a p53-regulated inhibitor of G₂-M progression. Mol. Cell, 1: 3-11, 1997.[Medline]
8. Marra, M. A., Kucaba, T. A., Hillier, L. W., and Waterston, R. H. High-throughput plasmid DNA purification for three cents per sample. Nucleic Acids Res., in press, 1999.
9. Chen H., Centola M., Altschul S. F., Metzger H. Characterization of gene expression in resting and activated mast cells. J. Exp. Med., 188: 1657-1668, 1998.[Abstract/Free Full Text]
10. Polyak K., Xia Y., Zweier J. L., Kinzler K. W., Vogelstein B. A model for p53-induced apoptosis. Nature (Lond.), 389: 300-305, 1997.[Medline]
11. Kleihues P., Burger P. C., Plate K. H., Ohgaki H., Cavenee W. K. Glioblastoma Kleihues P. Cavenee W. K. eds. . Pathology and Genetics: Tumors of the Nervous System, : IARC Lyon, France 1997.
12. Kleihues P., Ohgaki H. Primary and secondary glioblastomas: from concept to clinical diagnosis. Neuro-Oncology, 1: 44-51, 1999.[Abstract]
13. Richman A. V., Balis G. A., Maniscalco J. E. Primary intracerebral tumor with mixed chondrosarcoma and glioblastoma-gliosarcoma or sarcoglioma?. J. Neuropathol. & Exp. Neurol., 39: 329-335, 1980.[Medline]
14. Sarmiento J., Ferrer I., Pons L., Ferrer E. Cerebral mixed tumour: osteo-condrosarcoma-glioblastoma multiforme. Acta Neurochir., 50: 335-341, 1979.[Medline]
15. Jensen R. L. Growth factor-mediated angiogenesis in the malignant progression of glial tumors: a review. Surg. Neurol., 49: 189-195, 1998.[Medline]
16. Polyak K., Waldman T., He T-C., Kinzler K. W., Vogelstein B. Genetic determinants of p53 induced apoptosis and growth arrest. Genes Dev., 10: 1945-1952, 1996.[Abstract/Free Full Text]
17. He T-C., S. A. Rago C., Hermeking H., Zawel L., da Costa L. T., Morin P. J., Vogelstein B., Kinzler K. W. Identification of c-MYC as a target of the APC pathway. Science (Washington DC), 281: 1509-1512, 1998.[Abstract/Free Full Text]
18. Madden S. L., Galella E. A., Zhu J., Bertelsen A. H., Beaudry G. A. SAGE transcript profiles for p53-dependent growth regulation. Oncogene, 15: 1079-1085, 1997.[Medline]
19. von Mehren M., Weiner L. M. Monoclonal antibody-based therapy. Curr. Opin. Oncol., 8: 493-498, 1996.[Medline]
20. Gilboa E., Nair S. K., Lyerly H. K. Immunotherapy of cancer with dendritic-cell-based vaccines. Cancer Immunol. Immunother., 46: 82-87, 1998.[Medline]

This article has been cited by other articles:

				J. H. Cho-Vega, S. Tsavachidis, K.-A. Do, J. Nakagawa, L. J. Medeiros, and T. J. McDonnell Dicarbonyl/L-Xylulose Reductase: A Potential Biomarker Identified by Laser-Capture Microdissection-Micro Serial Analysis of Gene Expression of Human Prostate Adenocarcinoma Cancer Epidemiol. Biomarkers Prev., December 1, 2007; 16(12): 2615 - 2622. [Abstract] [Full Text] [PDF]

				M. D. Robinson and G. K. Smyth Moderated statistical tests for assessing differences in tag abundance Bioinformatics, November 1, 2007; 23(21): 2881 - 2887. [Abstract] [Full Text] [PDF]

				J. M. Cerutti, G. Oler, P. Michaluart Jr., R. Delcelo, R. M. Beaty, J. Shoemaker, and G. J. Riggins Molecular Profiling of Matched Samples Identifies Biomarkers of Papillary Thyroid Carcinoma Lymph Node Metastasis Cancer Res., August 15, 2007; 67(16): 7885 - 7892. [Abstract] [Full Text] [PDF]

				J. Degenhardt, M. Haubrock, J. Donitz, E. Wingender, and T. Crass DEEP--A tool for differential expression effector prediction Nucleic Acids Res., July 13, 2007; 35(suppl_2): W619 - W624. [Abstract] [Full Text] [PDF]

				K. Johansson, K. Ahlen, R. Rinaldi, K. Sahlander, A. Siritantikorn, and R. Morgenstern Microsomal glutathione transferase 1 in anticancer drug resistance Carcinogenesis, February 1, 2007; 28(2): 465 - 470. [Abstract] [Full Text] [PDF]

				M. Spitzner, J. Ousingsawat, K. Scheidt, K. Kunzelmann, and R. Schreiber Voltage-gated K+ channels support proliferation of colonic carcinoma cells FASEB J, January 1, 2007; 21(1): 35 - 44. [Abstract] [Full Text] [PDF]

				K. Nakayama, N. Nakayama, B. Davidson, J. J.-C. Sheu, N. Jinawath, A. Santillan, R. Salani, R. E. Bristow, P. J. Morin, R. J. Kurman, et al. A BTB/POZ protein, NAC-1, is related to tumor recurrence and is essential for tumor growth and survival PNAS, December 5, 2006; 103(49): 18739 - 18744. [Abstract] [Full Text] [PDF]

				K. F. Rodriguez, L. A. Blomberg, K. A. Zuelke, J. R. Miles, J. E. Alexander, and C. E. Farin Identification of candidate mRNAs associated with gonadotropin-induced maturation of murine cumulus oocyte complexes using serial analysis of gene expression Physiol Genomics, November 21, 2006; 27(3): 318 - 327. [Abstract] [Full Text] [PDF]

				P. Modena, E. Lualdi, F. Facchinetti, J. Veltman, J. F. Reid, S. Minardi, I. Janssen, F. Giangaspero, M. Forni, G. Finocchiaro, et al. Identification of Tumor-Specific Molecular Signatures in Intracranial Ependymoma and Association With Clinical Characteristics J. Clin. Oncol., November 20, 2006; 24(33): 5223 - 5233. [Abstract] [Full Text] [PDF]

				X. Ge, Q. Wu, Y.-C. Jung, J. Chen, and S. M. Wang A large quantity of novel human antisense transcripts detected by LongSAGE Bioinformatics, October 15, 2006; 22(20): 2475 - 2479. [Abstract] [Full Text] [PDF]

				C. E. Pelloski, E Lin, L. Zhang, W.K. A. Yung, H. Colman, J.-L. Liu, S. Y. Woo, A. B. Heimberger, D. Suki, M. Prados, et al. Prognostic Associations of Activated Mitogen-Activated Protein Kinase and Akt Pathways in Glioblastoma. Clin. Cancer Res., July 1, 2006; 12(13): 3935 - 3941. [Abstract] [Full Text] [PDF]

				L. Guo, Y. Ma, R. Ward, V. Castranova, X. Shi, and Y. Qian Constructing Molecular Classifiers for the Accurate Prognosis of Lung Adenocarcinoma. Clin. Cancer Res., June 1, 2006; 12(11): 3344 - 3354. [Abstract] [Full Text] [PDF]

				X.-H. Ma, S.-J. Hu, H. Ni, Y.-C. Zhao, Z. Tian, J.-L. Liu, G. Ren, X.-H. Liang, H. Yu, P. Wan, et al. Serial Analysis of Gene Expression in Mouse Uterus at the Implantation Site J. Biol. Chem., April 7, 2006; 281(14): 9351 - 9360. [Abstract] [Full Text] [PDF]

				C. M. Banwell, D. P. MacCartney, M. Guy, A. E. Miles, M. R. Uskokovic, J. Mansi, P. M. Stewart, L. P. O'Neill, B. M. Turner, K. W. Colston, et al. Altered nuclear receptor corepressor expression attenuates vitamin d receptor signaling in breast cancer cells. Clin. Cancer Res., April 1, 2006; 12(7): 2004 - 2013. [Abstract] [Full Text] [PDF]

				J. S. Johansen, B. V. Jensen, A. Roslind, D. Nielsen, and P. A. Price Serum YKL-40, A New Prognostic Biomarker in Cancer Patients? Cancer Epidemiol. Biomarkers Prev., February 1, 2006; 15(2): 194 - 202. [Abstract] [Full Text] [PDF]

				M. J. Yen, C.-Y. Hsu, T.-L. Mao, T-C. Wu, R. Roden, T.-L. Wang, and I.-M. Shih Diffuse Mesothelin Expression Correlates with Prolonged Patient Survival in Ovarian Serous Carcinoma Clin. Cancer Res., February 1, 2006; 12(3): 827 - 831. [Abstract] [Full Text] [PDF]

				X. Chen, J.-m. Wu, K. Hornischer, A. Kel, and E. Wingender TiProD: the Tissue-specific Promoter Database Nucleic Acids Res., January 1, 2006; 34(suppl_1): D104 - D107. [Abstract] [Full Text] [PDF]

				M. Hu and R. A. Shivdasani Overlapping Gene Expression in Fetal Mouse Intestine Development and Human Colorectal Cancer Cancer Res., October 1, 2005; 65(19): 8715 - 8722. [Abstract] [Full Text] [PDF]

				M. Roy, Q. Xu, and C. Lee Evidence that public database records for many cancer-associated genes reflect a splice form found in tumors and lack normal splice forms Nucleic Acids Res., September 7, 2005; 33(16): 5026 - 5033. [Abstract] [Full Text] [PDF]

				X. Wu, M. G. Walker, J. Luo, and L. Wei GBA server: EST-based digital gene expression profiling Nucleic Acids Res., July 1, 2005; 33(suppl_2): W673 - W676. [Abstract] [Full Text] [PDF]

				J. Pylouster, C. Senamaud-Beaufort, and T. E. Saison-Behmoaras WEBSAGE: a web tool for visual analysis of differentially expressed human SAGE tags Nucleic Acids Res., July 1, 2005; 33(suppl_2): W693 - W695. [Abstract] [Full Text] [PDF]

				J. N. Rich, C. Hans, B. Jones, E. S. Iversen, R. E. McLendon, B.K. A. Rasheed, A. Dobra, H. K. Dressman, D. D. Bigner, J. R. Nevins, et al. Gene Expression Profiling and Genetic Markers in Glioblastoma Survival Cancer Res., May 15, 2005; 65(10): 4051 - 4058. [Abstract] [Full Text] [PDF]

				T. W. Vogel, Z. Zhuang, J. Li, H. Okamoto, M. Furuta, Y.-S. Lee, W. Zeng, E. H. Oldfield, A. O. Vortmeyer, and R. J. Weil Proteins and Protein Pattern Differences between Glioma Cell Lines and Glioblastoma Multiforme Clin. Cancer Res., May 15, 2005; 11(10): 3624 - 3632. [Abstract] [Full Text] [PDF]

				I. T. Tai, M. Dai, and L. B. Chen Periostin induction in tumor cell line explants and inhibition of in vitro cell growth by anti-periostin antibodies Carcinogenesis, May 1, 2005; 26(5): 908 - 915. [Abstract] [Full Text] [PDF]

				B. Lin, J. T. White, W. Lu, T. Xie, A. G. Utleg, X. Yan, E. C. Yi, P. Shannon, I. Khrebtukova, P. H. Lange, et al. Evidence for the Presence of Disease-Perturbed Networks in Prostate Cancer Cells by Genomic and Proteomic Analyses: A Systems Approach to Disease Cancer Res., April 15, 2005; 65(8): 3081 - 3091. [Abstract] [Full Text] [PDF]

				V. De Corte, K. Van Impe, E. Bruyneel, C. Boucherie, M. Mareel, J. Vandekerckhove, and J. Gettemans Increased importin-{beta}-dependent nuclear import of the actin modulating protein CapG promotes cell invasion J. Cell Sci., October 15, 2004; 117(22): 5283 - 5292. [Abstract] [Full Text] [PDF]

				S. M. Hewitt, J. Dear, and R. A. Star Discovery of Protein Biomarkers for Renal Diseases J. Am. Soc. Nephrol., July 1, 2004; 15(7): 1677 - 1689. [Abstract] [Full Text] [PDF]

				J. M. Cuezva, G. Chen, A. M. Alonso, A. Isidoro, D. E. Misek, S. M. Hanash, and D. G. Beer The bioenergetic signature of lung adenocarcinomas is a molecular marker of cancer diagnosis and prognosis Carcinogenesis, July 1, 2004; 25(7): 1157 - 1163. [Abstract] [Full Text] [PDF]

				W. Ning, T. J. Chu, C. J. Li, A. M. K. Choi, and D. G. Peters Genome-wide analysis of the endothelial transcriptome under short-term chronic hypoxia Physiol Genomics, June 17, 2004; 18(1): 70 - 78. [Abstract] [Full Text] [PDF]

				R. Shao, S. Bao, X. Bai, C. Blanchette, R. M. Anderson, T. Dang, M. L. Gishizky, J. R. Marks, and X.-F. Wang Acquired Expression of Periostin by Human Breast Cancers Promotes Tumor Angiogenesis through Up-Regulation of Vascular Endothelial Growth Factor Receptor 2 Expression Mol. Cell. Biol., May 1, 2004; 24(9): 3992 - 4003. [Abstract] [Full Text] [PDF]

				N. Oue, Y. Hamai, Y. Mitani, S. Matsumura, Y. Oshimo, P. P. Aung, K. Kuraoka, H. Nakayama, and W. Yasui Gene Expression Profile of Gastric Carcinoma: Identification of Genes and Tags Potentially Involved in Invasion, Metastasis, and Carcinogenesis by Serial Analysis of Gene Expression Cancer Res., April 1, 2004; 64(7): 2397 - 2405. [Abstract] [Full Text] [PDF]

				M. Richards, S.-P. Tan, J.-H. Tan, W.-K. Chan, and A. Bongso The Transcriptome Profile of Human Embryonic Stem Cells as Defined by SAGE Stem Cells, January 1, 2004; 22(1): 51 - 64. [Abstract] [Full Text] [PDF]