 |
 |
GEO help: Mouse over screen elements for information. |
|
| Status |
Public on Jul 01, 2010 |
| Title |
Genome-wide analysis of in-trans correlations between copy number and expression with application to human breast cancer |
| Organism |
Homo sapiens |
| Experiment type |
Genome variation profiling by array
|
| Summary |
Background: Genomic instability is a hallmark of cancer and is frequently reflected in the DNA copy number profile and the expression profile of genes. Methodology is needed to compare array-CGH data and microarray expression data in order to unveil effects of DNA copy number alterations of a gene on the expression of the gene itself (in-cis effects) and on other genes (in-trans effects).
Results: We present a method for detection of gene dosage effects on gene expression that first searches for genome-wide in-cis associations between copy number and expression, and then investigates in-trans effects of these genes. A key element of the method is a correction step to eliminate potential confounding effects of in-cis associations when searching for in-trans associations. This facilitates reliable detection of local in-trans associations (i.e. associations within a chromosome) and alleviates the problem of co-occurring copy number aberrations across the genome. Applying the method to data collected from breast carcinomas from 20 women, we found 422 genes with expression levels that are highly correlated to DNA copy number (in-cis) in certain regions of the genome. Of these, 109 (26%) had significant trans-correlation with the expression levels of a large number of genes (20 or more). The strongest trans-regulators were also differentially expressed in subgroups that were found on the basis of expression data by classification of the tumors to previously defined breast cancer subtypes. A study of the most significant in-trans correlated genes yields significant enrichment of certain cancer related pathways and GO terms.
Conclusions: We introduce a method for the joint comparative analysis of expression and copy number profiles for the discovery and investigation of in-cis and in-trans regulatory mechanisms. Using data from 20 primary human breast cancers, a number of significant in-trans regulatory genes were found and their biological relevance indicated, calling for the application of the methods on larger data sets.
Keywords: Correlations between copy number and gene expression, breast cancer
|
| |
|
| Overall design |
Patient samples:
From May 1995 to December 1998, 920 patients referred for surgical treatment of breast cancer from five different hospitals in the Oslo region entered into a large study on detection and evaluation of isolated tumor cells in bone marrow. From theses 920 patients, fresh frozen tumor tissue samples from 20 patients were selected for this study. All 20 breast carcinomas contained >40% tumor cells, the majority of the tumor specimens represent tumor size T1/T2, node status N0/N1 (9/11) and histological grade 2 or 3. The 20 samples have been classified into five clinically relevant molecular tumor subclasses, based on their expression profile. Each sample was analyzed on both CGH- and gene expression microarrays (no replicates). All experiments were conducted at The Norwegian Radium Hospital.
Array CGH experiments:
Agilent's Human Genome CGH Microarray 44A contains 44,255 in situ synthesized 60-mer probes (3,877 controls) designed for studying copy number changes and representing most of the known or predicted human genes. The probes are, according to manufacturers description, enriched for cancer relevant genes representing both coding and non-coding sequences on the chromosomes.
Gene expression experiments:
Agilent Whole Human Genome Oligo Microarrays containing ~44,000 oligonucleotide probes were used for the gene expression experiments, according to manufacturer’s protocol.
|
| |
|
| Contributor(s) |
Baumbusch LO, Lipson D, Aarøe JM, Tsalenko A, Naume B, Liestøl K, Yakhini Z, Lingjærde OC, Børresen-Dale A |
| Citation missing |
Has this study been published? Please login to update or notify GEO. |
| |
| Submission date |
Feb 20, 2008 |
| Last update date |
Dec 06, 2012 |
| Contact name |
Jørgen Mømb Aarøe |
| E-mail(s) |
jorgen.aaroe@rr-research.no
|
| Phone |
+4791774653
|
| Fax |
+4722934440
|
| Organization name |
The Norwegian Radium Hospital
|
| Department |
Genetics
|
| Lab |
Genetics
|
| Street address |
Montebello
|
| City |
Oslo |
| ZIP/Postal code |
N-0310 |
| Country |
Norway |
| |
|
| Platforms (1) |
| GPL2873 |
Agilent-012750 Human Genome CGH Microarray 44A (Feature number version) |
|
| Samples (20)
|
|
| Relations |
| BioProject |
PRJNA107849 |
| Supplementary file |
Size |
Download |
File type/resource |
| GSE10583_RAW.tar |
198.6 Mb |
(http)(custom) |
TAR (of TXT) |
| Processed data included within Sample table |
|
|
|
|
 |
Less...
More...