Volume 16 Supplement 7
Co-modulation analysis of gene regulation in breast cancer reveals complex interplay between ESR1 and ERBB2 genes
- Yu-Chiao Chiu†1, 2,
- Chin-Ting Wu†1,
- Tzu-Hung Hsiao3,
- Yi-Pin Lai4,
- Chuhsing Kate Hsiao4, 5,
- Yidong Chen2, 6Email author and
- Eric Y Chuang1, 4Email author
© Chiu et al.; licensee BioMed Central Ltd. 2015
Published: 11 June 2015
Gene regulation is dynamic across cellular conditions and disease subtypes. From the aspect of regulation under modulation, regulation strength between a pair of genes can be modulated by (dependent on) expression abundance of another gene (modulator gene). Previous studies have demonstrated the involvement of genes modulated by single modulator genes in cancers, including breast cancer. However, analysis of multi-modulator co-modulation that can further delineate the landscape of complex gene regulation is, to our knowledge, unexplored previously. In the present study we aim to explore the joint effects of multiple modulator genes in modulating global gene regulation and dissect the biological functions in breast cancer.
To carry out the analysis, we proposed the Covariability-based Multiple Regression (CoMRe) method. The method is mainly built on a multiple regression model that takes expression levels of multiple modulators as inputs and regulation strength between genes as output. Pairs of genes were divided into groups based on their co-modulation patterns. Analyzing gene expression profiles from 286 breast cancer patients, CoMRe investigated ten candidate modulator genes that interacted and jointly determined global gene regulation. Among the candidate modulators, ESR1, ERBB2, and ADAM12 were found modulating the most numbers of gene pairs. The largest group of gene pairs was composed of ones that were modulated by merely ESR1. Functional annotation revealed that the group was significantly related to tumorigenesis and estrogen signaling in breast cancer. ESR1−ERBB2 co-modulation was the largest group modulated by more than one modulators. Similarly, the group was functionally associated with hormone stimulus, suggesting that functions of the two modulators are performed, at least partially, through modulation. The findings were validated in majorities of patients (> 99%) of two independent breast cancer datasets.
We have showed CoMRe is a robust method to discover critical modulators in gene regulatory networks, and it is capable of achieving reproducible and biologically meaningful results. Our data reveal that gene regulatory networks modulated by single modulator or co-modulated by multiple modulators play important roles in breast cancer. Findings of this report illuminate complex and dynamic gene regulation under modulation and its involvement in breast cancer.
With the advances in DNA microarray and the Next-Generation Sequencing (NGS) technologies, transcriptomic profiling of biological samples can be obtained fast and cost effectively. The high-throughput genomic data enable systematic inference of gene regulatory networks (GRNs) [1, 2]. In parallel, online databases, such as the Kyoto Encyclopedia of Genes and Genomes (KEGG)  and the Pathway Interaction Database (PID) , curate large volume of biologically (experimentally) validated gene regulatory pairs. These GRNs and pathways provide overall landscape of complex genome-wide gene regulation in biological systems. However, these gene regulatory relationships are typically derived under a single condition in a single cell line/tissue. From biological intuition, cells undergoing changes in cell cycle, environment, or cellular stress, and cells of different disease types or disease subtypes may recruit differential signaling pathways in response of cellular stimulation. Thus, strength and relationships of gene regulation are less likely to remain constitutive (unchanged) among these cells (reviewed in ). Ideker and Krogan proposed the scenario of "differential network biology" where GRNs and pathways can be massively rewired during adaptive cellular responses . Notably, dynamic interaction among proteins was shown to be predictive of breast cancer outcome , implying that studying the dynamic changes in network topology, as the differentially expressed genes, can provide biological clues of complex diseases.
From the viewpoint of regulation under modulation, the dynamics of cellular conditions can be determined (modulated) by status of certain modulator genes. In other words, gene A regulates gene B under the modulation of C refers to the scenario where regulation strength between gene pair A and B is dependent on expression level of the modulator C. For instance, previous study identified genes that were predictive of patient prognosis of lung adenocarcinomas in the RAS signature dependent manner . Also, competing endogenous RNA (ceRNA) regulation, referring to genes sharing common targeting miRNA that can regulate each other by competing for the limited pool of miRNAs [8–10], was shown to be modulated by expression levels of the common targeting miRNAs [11, 12]. In breast cancer, Estrogen Receptor (ER) is the most well studied modulator in gene regulation. Topological and temporal changes in GRN of transcription factors were observed in MCF7 breast cancer cell line upon estradiol stimulation . Furthermore, the ER encoding gene ESR1 was shown to be capable of modulating coexpression among a handful of genes . In order to systematically investigate gene regulation modulated by individual modulator genes, comprehensive mathematical methods were developed and carried out biologically testable findings [9, 15].
Gene regulation under modulation provides an alternative layer of gene regulatory networks. However, since gene regulation involves complex mechanism, especially in cancer, analysis based on individual modulator genes may be limited in understanding joint effects among multiple modulators and unveiling the landscape of modulation. Addressing this, in the present study we investigated the joint (cooperative, uncooperative, or dominant) effects of modulator genes in determining genome-wide gene regulation strength. Here we propose the Covariability-based Multiple Regression (CoMRe) method to model the relationships between multiple modulator genes and modulated gene-gene regulation in breast cancer. CoMRe was built mainly based on the multiple regression analysis which takes expression levels of modulators as model inputs and strength of gene-gene regulation, measured by our developed parameter "covariability", as output. On the other hand, investigation into functions governed by gene modulation in breast cancer remains largely unexplored. Thus, we further analyzed and interpreted the results identified by CoMRe in the systematic functional level. Collectively, the present study is aimed to statistically infer the relationship between multiple modulators and modulated gene regulation and to study the associated biological functions in breast cancer.
Results and discussion
Model overview of CoMRe
List of the 10 candidate modulator genes.
Entrez gene name
ADAM metallopeptidase domain 12
chemokine (C-C motif) ligand 5
v-erb-b2 avian erythroblastic leukemia viral oncogene homolog 2
estrogen receptor 1
ligand-dependent nuclear receptor
insulin-like growth factor 1 (somatomedin C)
macrophage migration inhibitory factor (glycosylation-inhibiting factor)
marker of proliferation Ki-67
v-myc avian myelocytomatosis viral oncogene homolog
reversion-inducing-cysteine-rich protein with kazal motifs
tumor protein p53
Dissecting individual effects of modulator genes in modulating gene regulation
Among the 10 candidate modulator genes, notably, the well-studied modulator gene ESR1 was found significantly modulating the most number of gene pairs (2,449,249 gene pairs, 17.39% of all pairs), followed by v-erb-b2 avian erythroblastic leukemia viral oncogene homolog 2 (ERBB2; 1,772,703 pairs, 12.59%) and ADAM metallopeptidase domain 12 (ADAM12; 1,764,441 pairs, 12.53%) (Figure 2E). Together with progesterone receptor (PR), ER and Her2 (ERBB2 encoded protein) are genes currently used for molecular subtyping of breast cancers. The results indicate that the two genes define distinct molecular characteristics in breast cancer partially through modulation of gene regulation. Among the modulator genes, reversion-inducing-cysteine-rich protein with kazal motifs (RECK), tumor protein p53 (TP53), and insulin-like growth factor 1 (IGF1), were found to modulate the least numbers of gene pairs (711,658 (5.05%), 998,142 (7.09%), and 1,113,258 (7.90%) gene pairs, respectively; Figure 2E). Although these genes are related to essential functions of breast tumor progression, they may possess relatively minor, or overtaken by other candidate modulators, effects in modulation of gene regulation. To generate a random baseline of our results, we replaced the inputs of modulator expression levels with ten randomly simulated variables and reran the analyses. Each of the ten random variables showed significance only in 3.33% to 5.81%, approximating the p-value cutoff of 0.05, of the 14,084,778 gene pairs. Taken together, our data suggest the capability of CoMRe in identifying both biologically well-known results and novel insights into other candidate modulator genes.
Investigating joint effects of multiple modulator genes in modulating gene regulation and related biological functions
Top co-modulation groups among the 10 candidate modulator genes.
Number of modulated gene pairs
Top hub genesb
Number of core modulated gene pairsc
Number of core modulated genesc
None of the 10 modulators
RALYL (4253); CALML3 (4095); KLK12 (3920)
NFIX (1092); SMPDL3A (1072); VEGFA (1045)
LUM (1005); MYST4 (993); CHST1 (988)
CTSW (1622); CCL4 (1209); CCR5 (1206)
CPS1 (1160); GDPD3 (929); KLHDC4 (914)
MED1 (1012); PGAP3 (1005); STARD3 (1000)
RPS2 (1313); FTL (1200); PRKCSH (1137)
BOLA1 (884); FOSB (844); SLC38A2 (817)
TP53 (1636); RAGE (1005); RABGAP1 (909)
PLA2G2A (672); TNXB (662); CIDEA (659)
RECK (1693); DNAH3 (1640); OPHN1 (1622)
MIA (780); KRT6B (631); FYCO1 (628)
AZI1 (753); AQP5 (688); C3ORF37 (515)
HOXA10 (539); PXDN (465); APOC2 (462)
CD38 (661); PRF1 (660); IL2RG (544)
KMO (534); ETFA (394); NAAA (362)
WNK1 (290); IGFBP4 (288); MFAP2 (279)
TTC23 (436); KIT (306); PRPF18 (280)
MAP7 (347); VCAM1 (270); TRAPPC10 (269)
FBN1 (259); MARCH5 (241); GADD45B (234)
CRLF1 (565); PGC (435); ITPKB (317)
Top 3 clusters of enriched GO molecular functions and biological processes in ESR1 modulated genes.
Number of genes
Cluster 1 (Enrichment Score: 4.23)
DNA metabolic process
7.85 × 10-6
Cellular response to stress
8.80 × 10-5
Response to DNA damage stimulus
1.32 × 10-4
1.33 × 10-4
Cluster 2 (Enrichment Score: 3.95)
Identical protein binding
4.33 × 10-5
Protein dimerization activity
1.48 × 10-4
Protein homodimerization activity
2.21 × 10-4
Cluster 3 (Enrichment Score: 3.66)
Response to estrogen stimulus
1.98 × 10-5
Response to organic substance
2.90 × 10-5
Response to steroid hormone stimulus
7.27 × 10-5
Response to hormone stimulus
3.94 × 10-4
Response to endogenous stimulus
6.91 × 10-4
Response to estradiol stimulus
Complex and tight interplay of ESR1 and ERBB2 modulation
Top 3 clusters of enriched GO molecular functions and biological processes in ESR1−ERBB2 co-modulated genes.
Number of genes
Cluster 1 (Enrichment Score: 2.22)
Response to hormone stimulus
6.49 × -04
Response to endogenous stimulus
Response to steroid hormone stimulus
Response to organic substance
Response to peptide hormone stimulus
Response to insulin stimulus
Response to estrogen stimulus
Cluster 2 (Enrichment Score: 1.92)
Electron carrier activity
Cluster 3 (Enrichment Score: 1.83)
Protein dimerization activity
Identical protein binding
Protein homodimerization activity
External validation of co-modulation patterns
In order to test the reproducibility and reliability of CoMRe among different cohorts, we analyzed two independent breast cancer datasets, GSE4922 and GSE25066, for validation. Based on the co-modulation patterns (β values of the modulator genes) obtained from GSE2034, we computed the "estimated" covariability profile for each patient in the two validation datasets using corresponding expression data of the modulator genes. The real covariability profiles were calculated using global gene expression data in each of the validation datasets. Notably, the estimated and real covariability profiles were significantly positively correlated (Pearson correlation p-value < 0.05) in 99.31% (287 out of 289, one-sample z-test p-value < computing precision of double-precision floating point, hereafter referred to as p-value ~0) and 99.80% (507 out of 508, p-value ~0) of patients in GSE4922 and GSE25066, respectively. Similarly, for ESR1−ERBB2 co-modulated gene pairs, the results were validated in 100% (all of 289 patients, p-value ~0) and 99.80% (507 out of 508, p-value ~0) of patients. The data suggest the stability of modulation effects among different cohorts and the reproducibility of results identified by CoMRe.
Limitations and future work
By far validation of modulator genes through biological experiments is very limited. In breast cancer, ER is the most well-studied modulator gene. In the present study, in addition to the ER encoding gene ESR1, we exploratorily included 9 more genes related to essential functions in breast tumor progression, with previously undiscovered function of modulation. Our data first validated the role of ESR1 as a modulator gene and suggested that it may jointly work with other modulators. Also, the results implied the existence of other modulators genes in breast cancer, such as ERBB2 and ADAM12. However, ~37% of gene pairs were not modulated by any of the 10 candidate modulators, suggestive of the need for inclusion of other modulators. We have demonstrated the performance of CoMRe and the benefits to study modulation in the joint manner. With advances in biological exploration of modulator genes, CoMRe can be employed to reveal more biologically meaningful findings.
Investigation of casual relationships between genes is one of the crucial topics in regulatory biology. Indeed, correlation coefficients, as well as mutual information, are not capable of measuring causal relationships between factors. However, analyses of modulated gene regulation typically focus on how expression levels of modulators affect regulating strength, instead of the causal relationships, between modulated genes. Previous studies have used non-causal statistical methods to reach comprehensive results in single modulator analyses [9, 15]. In this study, our objective is to extend the analysis to inferring multiple modulators co-modulated gene regulation, using a correlation-based regression approach. Therefore, CoMRe was designed to evaluate how co-variability of genome-wide gene pairs was dependent on modulator genes based on a multiple regression model; the analysis was focused on modulation, rather than direct or causal regulation, or co-regulation (i.e., regulated changes in gene expression levels).
CoMRe is built on the basis of a multiple linear regression model. In statistics, multiple regression analysis typically assumes the independence among input variables (i.e., expression profiles of modulator genes of CoMRe). However, biological intuition is that two genes can hardly be independent to each other in cells. In previous studies, multiple regression model has been widely utilized to study genes [31, 32], different data types (from gene expression, transcription factor binding, and drug response data) [31, 33, 34], and survival significance of multiple genomic features (clinical subtypes and prognostic factors) [35, 36]. Findings of these reports suggest that multiple regression can achieve biologically meaningful results, in spite of the moderate dependency of genomic features. Thus, we followed these literatures and designed CoMRe to study multi-modulator modulation. Future efforts may be spent on developing algorithms that can take dependent genomic features and enable statistically more meaningful inference.
In the present study, we presented the CoMRe algorithm for systematically investigating how multiple modulator genes jointly determine pairwise regulation strength of modulated genes. The algorithm was designed based on a multiple regression model for gene-gene covariability that measures how two genes regulate each other in each patient. Among the ten candidate modulator genes, the positive control ESR1 and two genes with essential functions in breast cancer were found modulating the most numbers of gene pairs. Through functional annotation analysis, we showed that genes modulated by merely single modulator or co-modulated by multiple modulators play important roles in breast cancer. We elucidate that ESR1 and ERBB2 share complex interplay between each other in the aspect of gene modulation. We also demonstrated that the co-modulation patterns are stably retained and the results identified by CoMRe are highly reproducible among different cohorts. From the viewpoint of multi-modulator modulation, this study paves the way for better understanding complex gene regulation in breast cancer.
We analyzed gene expression profiles of 286 lymph-node negative breast cancers, of which 180 were relapse-free patients and 106 developed distant metastasis, from GSE2034 . The samples were profiled with Affymetrix Human Genome U133A Arrays. We reprocessed the raw intensity values of CEL files using the Robust Microarray Analysis (RMA) algorithm into log-2 scaled probe set level expression levels. For multiple probe sets representing one unique gene, the one with the largest coefficient of variation (CV) was selected as the representative probe set. To eliminate computationally non-informative and background probe sets, probe sets with CV values < 5% or average expression levels < 6 (in the log-2 scale) across samples were filtered out from subsequent analysis.
We included two independent gene expression datasets for validation, composed of primary invasive breast tumors (NCBI/GEO Accession Number GSE4922 ) from Uppsala, Stockholm, and Singapore cohorts and pre-treatment invasive breast cancer patients in M. D. Anderson Cancer Center (GSE25066 [39, 40]). 289 and 508 samples in the two datasets with complete molecular and clinical information were analyzed. The datasets were profiled with Affymetrix Human Genome U133A Arrays and we reprocessed the microarray data following identical procedures as described above. For each of the genes selected for analysis in GSE2034, one probe with the largest CV value in each validation dataset was extracted from the validation dataset for analysis.
Covariability-based multiple regression
where denotes the expression level of gene i in patient k, and and represent the average and standard deviation of gene i across patients. Denotation of gene j is identical. The covariability was designed to measure the magnitude of changes in two genes in the same direction in one sample. Mathematically, it is simply the per-sample product-moment component in the calculation of Pearson correlation coefficient ρ; i.e., .
where the regressand is the covariability vector of gene i and j, regressor denotes the expression profile of modulator gene m, represents regression coefficients for modulator gene m, and is the error vector. Statistical significance of the obtained regression coefficients was assessed using t-test. The regression model was iteratively applied to each combination of gene i and j in the microarray dataset. Thus, for each gene pair i and j, each modulator gene m takes a regression β value and p-value. A significant p-value indicates that the modulator is significantly predictive of the covariability (i.e., regulation strength) of corresponding gene pair. We defined the co-modulation patterns for each gene pair as the M-length vectors of β values and p-values for M modulator genes. To further dissecting the gene pairs based on their co-modulation patterns, we grouped gene pairs that were significantly modulated by the same set of modulators.
Statistical analyses and functional annotation analysis
Fisher's exact test was employed to infer the significance of co-occurrence of significant modulator genes in the co-modulation patterns. Also, given a sample proportion , we estimated the 95% confidence interval of the population proportion by , where N denotes the sample size. To gain biological insights, we utilized the Database for Annotation, Visualization and Integrated Discovery (DAVID) v6.7 web tool [41, 42] to identify the Gene Ontology (GO) [43, 44] biological process and molecular function terms that exhibit significant enrichment in our gene list. In order to interpret the results in a more systematic and comprehensive level, we grouped highly overlapped GO terms into clusters using the DAVID Functional Annotation Clustering tool.
The study is partly supported by the Ministry of Science and Technology of Taiwan (grant ID 103-2917-I-002-166). The authors also wish to thank Center of Genomic Medicine, National Taiwan University for financial support and computing servers. The study is also partially supported by NCI grant (1R01CA152063-02) and Greehey Children's Cancer Research Institute (GCCRI) intramural research fund. The authors also greatly appreciate the brilliant and constructive inputs from reviewers and participants of the International Conference on Intelligent Biology and Medicine (ICIBM 2014).
The publication costs for this article were funded by the Greehey Children's Cancer Research Institute's intramural research fund.
This article has been published as part of BMC Genomics Volume 16 Supplement 7, 2015: Selected articles from The International Conference on Intelligent Biology and Medicine (ICIBM) 2014: Genomics. The full contents of the supplement are available online at http://www.biomedcentral.com/bmcgenomics/supplements/16/S7.
- Teichmann SA, Babu MM: Gene regulatory network growth by duplication. Nature genetics. 2004, 36 (5): 492-496. 10.1038/ng1340.View ArticlePubMedGoogle Scholar
- Schaefer CF, Anthony K, Krupa S, Buchoff J, Day M, Hannay T, Buetow KH: PID: the Pathway Interaction Database. Nucleic acids research. 2009, 37 (Database): D674-679. 10.1093/nar/gkn653.PubMed CentralView ArticlePubMedGoogle Scholar
- Kanehisa M, Goto S: KEGG: kyoto encyclopedia of genes and genomes. Nucleic acids research. 2000, 28 (1): 27-30. 10.1093/nar/28.1.27.PubMed CentralView ArticlePubMedGoogle Scholar
- Li KC: Genome-wide coexpression dynamics: theory and application. Proceedings of the National Academy of Sciences of the United States of America. 2002, 99 (26): 16875-16880. 10.1073/pnas.252466999.PubMed CentralView ArticlePubMedGoogle Scholar
- Ideker T, Krogan NJ: Differential network biology. Molecular systems biology. 2012, 8: 565-PubMed CentralView ArticlePubMedGoogle Scholar
- Taylor IW, Linding R, Warde-Farley D, Liu Y, Pesquita C, Faria D, Bull S, Pawson T, Morris Q, Wrana JL: Dynamic modularity in protein interaction networks predicts breast cancer outcome. Nat Biotechnol. 2009, 27 (2): 199-204. 10.1038/nbt.1522.View ArticlePubMedGoogle Scholar
- Luo J, Emanuele MJ, Li D, Creighton CJ, Schlabach MR, Westbrook TF, Wong KK, Elledge SJ: A genome-wide RNAi screen identifies multiple synthetic lethal interactions with the Ras oncogene. Cell. 2009, 137 (5): 835-848. 10.1016/j.cell.2009.05.006.PubMed CentralView ArticlePubMedGoogle Scholar
- Karreth FA, Tay Y, Perna D, Ala U, Tan SM, Rust AG, DeNicola G, Webster KA, Weiss D, Perez-Mancera PA, et al: In vivo identification of tumor- suppressive PTEN ceRNAs in an oncogenic BRAF-induced mouse model of melanoma. Cell. 2011, 147 (2): 382-395. 10.1016/j.cell.2011.09.032.PubMed CentralView ArticlePubMedGoogle Scholar
- Sumazin P, Yang X, Chiu HS, Chung WJ, Iyer A, Llobet-Navas D, Rajbhandari P, Bansal M, Guarnieri P, Silva J, et al: An extensive microRNA-mediated network of RNA-RNA interactions regulates established oncogenic pathways in glioblastoma. Cell. 2011, 147 (2): 370-381. 10.1016/j.cell.2011.09.041.PubMed CentralView ArticlePubMedGoogle Scholar
- Tay Y, Kats L, Salmena L, Weiss D, Tan SM, Ala U, Karreth F, Poliseno L, Provero P, Di Cunto F, et al: Coding-independent regulation of the tumor suppressor PTEN by competing endogenous mRNAs. Cell. 2011, 147 (2): 344-357. 10.1016/j.cell.2011.09.029.PubMed CentralView ArticlePubMedGoogle Scholar
- Ala U, Karreth FA, Bosia C, Pagnani A, Taulli R, Leopold V, Tay Y, Provero P, Zecchina R, Pandolfi PP: Integrated transcriptional and competitive endogenous RNA networks are cross-regulated in permissive molecular environments. Proceedings of the National Academy of Sciences of the United States of America. 2013, 110 (18): 7154-7159. 10.1073/pnas.1222509110.PubMed CentralView ArticlePubMedGoogle Scholar
- Chiu YC, Hsiao TH, Chen Y, Chuang EY: Parameter optimization for constructing competing endogenous RNA regulatory network in glioblastoma multiforme and other cancers. BMC genomics. 2015, 6 (Suppl 4): S1-View ArticleGoogle Scholar
- Shen C, Huang Y, Liu Y, Wang G, Zhao Y, Wang Z, Teng M, Wang Y, Flockhart DA, Skaar TC, et al: A modulated empirical Bayes model for identifying topological and temporal estrogen receptor alpha regulatory networks in breast cancer. BMC systems biology. 2011, 5: 67-10.1186/1752-0509-5-67.PubMed CentralView ArticlePubMedGoogle Scholar
- Wilson CA, Dering J: Recent translational research: microarray expression profiling of breast cancer--beyond classification and prognostic markers?. Breast cancer research : BCR. 2004, 6 (5): 192-200. 10.1186/bcr917.PubMed CentralView ArticlePubMedGoogle Scholar
- Wang K, Saito M, Bisikirska BC, Alvarez MJ, Lim WK, Rajbhandari P, Shen Q, Nemenman I, Basso K, Margolin AA, et al: Genome-wide identification of post-translational modulators of transcription factor activity in human B cells. Nature biotechnology. 2009, 27 (9): 829-839. 10.1038/nbt.1563.PubMed CentralView ArticlePubMedGoogle Scholar
- Badve S, Turbin D, Thorat MA, Morimiya A, Nielsen TO, Perou CM, Dunn S, Huntsman DG, Nakshatri H: FOXA1 expression in breast cancer--correlation with luminal subtype A and survival. Clinical cancer research : an official journal of the American Association for Cancer Research. 2007, 13 (15 Pt 1): 4415-4421.View ArticleGoogle Scholar
- Habashy HO, Powe DG, Rakha EA, Ball G, Paish C, Gee J, Nicholson RI, Ellis IO: Forkhead-box A1 (FOXA1) expression in breast cancer and its prognostic significance. European journal of cancer. 2008, 44 (11): 1541-1551. 10.1016/j.ejca.2008.04.020.View ArticlePubMedGoogle Scholar
- Thorat MA, Marchio C, Morimiya A, Savage K, Nakshatri H, Reis-Filho JS, Badve S: Forkhead box A1 expression in breast cancer is associated with luminal subtype and good prognosis. Journal of clinical pathology. 2008, 61 (3): 327-332.View ArticlePubMedGoogle Scholar
- Dong C, Yuan T, Wu Y, Wang Y, Fan TW, Miriyala S, Lin Y, Yao J, Shi J, Kang T, et al: Loss of FBP1 by Snail-mediated repression provides metabolic advantages in basal-like breast cancer. Cancer cell. 2013, 23 (3): 316-331. 10.1016/j.ccr.2013.01.022.PubMed CentralView ArticlePubMedGoogle Scholar
- van 't Veer LJ, Dai H, van de Vijver MJ, He YD, Hart AA, Mao M, Peterse HL, van der Kooy K, Marton MJ, Witteveen AT, et al: Gene expression profiling predicts clinical outcome of breast cancer. Nature. 2002, 415 (6871): 530-536. 10.1038/415530a.View ArticlePubMedGoogle Scholar
- Leo JC, Wang SM, Guo CH, Aw SE, Zhao Y, Li JM, Hui KM, Lin VC: Gene regulation profile reveals consistent anticancer properties of progesterone in hormone-independent breast cancer cells transfected with progesterone receptor. International journal of cancer Journal international du cancer. 2005, 117 (4): 561-568. 10.1002/ijc.21186.View ArticlePubMedGoogle Scholar
- Borowsky AD, Namba R, Young LJ, Hunter KW, Hodgson JG, Tepper CG, McGoldrick ET, Muller WJ, Cardiff RD, Gregg JP: Syngeneic mouse mammary carcinoma cell lines: two closely related cell lines with divergent metastatic behavior. Clinical & experimental metastasis. 2005, 22 (1): 47-59. 10.1007/s10585-005-2908-5.View ArticleGoogle Scholar
- Ryden L, Linderholm B, Nielsen NH, Emdin S, Jonsson PE, Landberg G: Tumor specific VEGF-A and VEGFR2/KDR protein are co-expressed in breast cancer. Breast cancer research and treatment. 2003, 82 (3): 147-154. 10.1023/B:BREA.0000004357.92232.cb.View ArticlePubMedGoogle Scholar
- Balasubramanian SP, Cox A, Cross SS, Higham SE, Brown NJ, Reed MW: Influence of VEGF-A gene variation and protein levels in breast cancer susceptibility and severity. International journal of cancer Journal international du cancer. 2007, 121 (5): 1009-1016. 10.1002/ijc.22772.View ArticlePubMedGoogle Scholar
- Khanna KK, Jackson SP: DNA double-strand breaks: signaling, repair and the cancer connection. Nature genetics. 2001, 27 (3): 247-254. 10.1038/85798.View ArticlePubMedGoogle Scholar
- Bartkova J, Horejsi Z, Koed K, Kramer A, Tort F, Zieger K, Guldberg P, Sehested M, Nesland JM, Lukas C, et al: DNA damage response as a candidate anti-cancer barrier in early human tumorigenesis. Nature. 2005, 434 (7035): 864-870. 10.1038/nature03482.View ArticlePubMedGoogle Scholar
- Bosserhoff AK, Kaufmann M, Kaluza B, Bartke I, Zirngibl H, Hein R, Stolz W, Buettner R: Melanoma-inhibiting activity, a novel serum marker for progression of malignant melanoma. Cancer research. 1997, 57 (15): 3149-3153.PubMedGoogle Scholar
- Auge JM, Molina R, Filella X, Bosch E, Gonzalez Cao M, Puig S, Malvehy J, Castel T, Ballesta AM: S-100beta and MIA in advanced melanoma in relation to prognostic factors. Anticancer research. 2005, 25 (3A): 1779-1782.PubMedGoogle Scholar
- Ponzo MG, Lesurf R, Petkiewicz S, O'Malley FP, Pinnaduwage D, Andrulis IL, Bull SB, Chughtai N, Zuo D, Souleimanova M, et al: Met induces mammary tumors with diverse histologies and is associated with poor outcome and human basal breast cancer. Proceedings of the National Academy of Sciences of the United States of America. 2009, 106 (31): 12903-12908. 10.1073/pnas.0810402106.PubMed CentralView ArticlePubMedGoogle Scholar
- Hynes NE, Lane HA: ERBB receptors and cancer: the complexity of targeted inhibitors. Nature reviews Cancer. 2005, 5 (5): 341-354. 10.1038/nrc1609.View ArticlePubMedGoogle Scholar
- Comings DE, Gade-Andavolu R, Gonzalez N, Wu S, Muhleman D, Blake H, Dietz G, Saucier G, MacMurray JP: Comparison of the role of dopamine, serotonin, and noradrenaline genes in ADHD, ODD and conduct disorder: multivariate regression analysis of 20 genes. Clinical genetics. 2000, 57 (3): 178-196.View ArticlePubMedGoogle Scholar
- Lamba V, Panetta JC, Strom S, Schuetz EG: Genetic predictors of interindividual variability in hepatic CYP3A4 expression. The Journal of pharmacology and experimental therapeutics. 2010, 332 (3): 1088-1099. 10.1124/jpet.109.160804.PubMed CentralView ArticlePubMedGoogle Scholar
- Kutalik Z, Beckmann JS, Bergmann S: A modular approach for integrative analysis of large-scale gene-expression and drug-response data. Nature biotechnology. 2008, 26 (5): 531-539. 10.1038/nbt1397.View ArticlePubMedGoogle Scholar
- Gao F, Foat BC, Bussemaker HJ: Defining transcriptional networks through integrative modeling of mRNA expression and transcription factor binding data. BMC bioinformatics. 2004, 5: 31-10.1186/1471-2105-5-31.PubMed CentralView ArticlePubMedGoogle Scholar
- Blows FM, Driver KE, Schmidt MK, Broeks A, van Leeuwen FE, Wesseling J, Cheang MC, Gelmon K, Nielsen TO, Blomqvist C, et al: Subtyping of breast cancer by immunohistochemistry to investigate a relationship between subtype and short and long term survival: a collaborative analysis of data for 10,159 cases from 12 studies. PLoS medicine. 2010, 7 (5): e1000279-10.1371/journal.pmed.1000279.PubMed CentralView ArticlePubMedGoogle Scholar
- Bedrosian I, Hu CY, Chang GJ: Population-based study of contralateral prophylactic mastectomy and survival outcomes of breast cancer patients. Journal of the National Cancer Institute. 2010, 102 (6): 401-409. 10.1093/jnci/djq018.PubMed CentralView ArticlePubMedGoogle Scholar
- Wang Y, Klijn JG, Zhang Y, Sieuwerts AM, Look MP, Yang F, Talantov D, Timmermans M, Meijer-van Gelder ME, Yu J, et al: Gene-expression profiles to predict distant metastasis of lymph-node-negative primary breast cancer. Lancet. 2005, 365 (9460): 671-679. 10.1016/S0140-6736(05)70933-8.View ArticlePubMedGoogle Scholar
- Ivshina AV, George J, Senko O, Mow B, Putti TC, Smeds J, Lindahl T, Pawitan Y, Hall P, Nordgren H, et al: Genetic reclassification of histologic grade delineates new clinical subtypes of breast cancer. Cancer research. 2006, 66 (21): 10292-10301. 10.1158/0008-5472.CAN-05-4414.View ArticlePubMedGoogle Scholar
- Hatzis C, Pusztai L, Valero V, Booser DJ, Esserman L, Lluch A, Vidaurre T, Holmes F, Souchon E, Wang H, et al: A genomic predictor of response and survival following taxane-anthracycline chemotherapy for invasive breast cancer. JAMA : the journal of the American Medical Association. 2011, 305 (18): 1873-1881. 10.1001/jama.2011.593.View ArticlePubMedGoogle Scholar
- Itoh M, Iwamoto T, Matsuoka J, Nogami T, Motoki T, Shien T, Taira N, Niikura N, Hayashi N, Ohtani S, et al: Estrogen receptor (ER) mRNA expression and molecular subtype distribution in ER-negative/progesterone receptor-positive breast cancers. Breast cancer research and treatment. 2014, 143 (2): 403-409. 10.1007/s10549-013-2763-z.View ArticlePubMedGoogle Scholar
- Huang da W, Sherman BT, Lempicki RA: Systematic and integrative analysis of large gene lists using DAVID bioinformatics resources. Nature protocols. 2009, 4 (1): 44-57.View ArticlePubMedGoogle Scholar
- Huang da W, Sherman BT, Lempicki RA: Bioinformatics enrichment tools: paths toward the comprehensive functional analysis of large gene lists. Nucleic acids research. 2009, 37 (1): 1-13. 10.1093/nar/gkn923.PubMed CentralView ArticlePubMedGoogle Scholar
- Ashburner M, Ball CA, Blake JA, Botstein D, Butler H, Cherry JM, Davis AP, Dolinski K, Dwight SS, Eppig JT, et al: Gene ontology: tool for the unification of biology. The Gene Ontology Consortium. Nature genetics. 2000, 25 (1): 25-29. 10.1038/75556.PubMed CentralView ArticlePubMedGoogle Scholar
- Harris MA, Clark J, Ireland A, Lomax J, Ashburner M, Foulger R, Eilbeck K, Lewis S, Marshall B, Mungall C, et al: The Gene Ontology (GO) database and informatics resource. Nucleic acids research. 2004, 32 (Database): D258-261.PubMedGoogle Scholar
This article is published under license to BioMed Central Ltd. This is an Open Access article distributed under the terms of the Creative Commons Attribution License (http://creativecommons.org/licenses/by/4.0), which permits unrestricted use, distribution, and reproduction in any medium, provided the original work is properly cited. The Creative Commons Public Domain Dedication waiver (http://creativecommons.org/publicdomain/zero/1.0/) applies to the data made available in this article, unless otherwise stated.