Dataset Information

Extracting disease risk profiles from expression data for linkage analysis: application to prostate cancer.

ABSTRACT: The genetic factors underlying many complex traits are not well understood. The Genetic Analysis Workshop 15 Problem 1 data present the opportunity to explore whether gene expression data from microarrays can be utilized to define useful phenotypes for linkage analysis in complex diseases. We utilize expression profiles for multiple genes that have been associated with a disease to develop a composite 'risk profile' that can be used to map other loci involved in the same disease process. Using prostate cancer as our disease of interest, we identified 26 genes whose expression levels had previously been associated with prostate cancer and defined three phenotypes: high, neutral, or low risk profiles, based on individual expression levels. Linkage analyses using MCLINK, a Markov-chain Monte Carlo method, and MERLIN were performed for all three phenotypes. Both methods were in very close agreement. Genome-wide suggestive linkage evidence was observed on chromosomes 6 and 4. It was interesting to note that the linkage signals did not appear to be strongly influenced by the location of the original 26 genes used in the phenotype definition, indicating that composite measures may have potential to locate additional genes in the same process. In this example, however, extreme caution is necessary in any extrapolation of the identified loci to prostate cancer due to the lack of data regarding the behavior of these genes' expression level in lymphoblastoid cells. Our results do indicate there exists potential to augment our current knowledge about the relationships among genes associated with complex diseases using expression data.

SUBMITTER: Christensen GB

PROVIDER: S-EPMC2367601 | biostudies-literature | 2007

REPOSITORIES: biostudies-literature

ACCESS DATA

Publications

Extracting disease risk profiles from expression data for linkage analysis: application to prostate cancer.

Christensen G Bryce GB Cannon-Albright Lisa A LA Thomas Alun A Camp Nicola J NJ

BMC proceedings 20071218

The genetic factors underlying many complex traits are not well understood. The Genetic Analysis Workshop 15 Problem 1 data present the opportunity to explore whether gene expression data from microarrays can be utilized to define useful phenotypes for linkage analysis in complex diseases. We utilize expression profiles for multiple genes that have been associated with a disease to develop a composite 'risk profile' that can be used to map other loci involved in the same disease process. Using p ...[more]

PMID: 18466585

Similar Datasets

Project description:BackgroundCells dynamically adapt their gene expression patterns in response to various stimuli. This response is orchestrated into a number of gene expression modules consisting of co-regulated genes. A growing pool of publicly available microarray datasets allows the identification of modules by monitoring expression changes over time. These time-series datasets can be searched for gene expression modules by one of the many clustering methods published to date. For an integrative analysis, several time-series datasets can be joined into a three-dimensional gene-condition-time dataset, to which standard clustering or biclustering methods are, however, not applicable. We thus devise a probabilistic clustering algorithm for gene-condition-time datasets.ResultsIn this work, we present the EDISA (Extended Dimension Iterative Signature Algorithm), a novel probabilistic clustering approach for 3D gene-condition-time datasets. Based on mathematical definitions of gene expression modules, the EDISA samples initial modules from the dataset which are then refined by removing genes and conditions until they comply with the module definition. A subsequent extension step ensures gene and condition maximality. We applied the algorithm to a synthetic dataset and were able to successfully recover the implanted modules over a range of background noise intensities. Analysis of microarray datasets has lead us to define three biologically relevant module types: 1) We found modules with independent response profiles to be the most prevalent ones. These modules comprise genes which are co-regulated under several conditions, yet with a different response pattern under each condition. 2) Coherent modules with similar responses under all conditions occurred frequently, too, and were often contained within these modules. 3) A third module type, which covers a response specific to a single condition was also detected, but rarely. All of these modules are essentially different types of biclusters.ConclusionWe successfully applied the EDISA to different 3D datasets. While previous studies were mostly aimed at detecting coherent modules only, our results show that coherent responses are often part of a more general module type with independent response profiles under different conditions. Our approach thus allows for a more comprehensive view of the gene expression response. After subsequent analysis of the resulting modules, the EDISA helped to shed light on the global organization of transcriptional control. An implementation of the algorithm is available at http://www-ra.informatik.uni-tuebingen.de/software/IAGEN/.

Project description:ObjectiveThis study sought whether higher risk patients with coronary heart disease (CHD) benefit more from intensive disease management.DesignLongitudinal cohort study.SettingState-wide public hospitals (Queensland, Australia).ParticipantsThis longitudinal study included 20 426 patients hospitalised in 2010 with CHD as the principal diagnosis. Patients were followed-up for 5 years.Primary and secondary outcomes and measuresThe primary outcome was days alive and out of hospital (DAOH) within 5 years of hospital discharge. Secondary outcomes included all-cause readmission and all-cause mortality. A previously developed and validated risk score (PEGASUS-TIMI54) was used to estimate the risk of secondary events. Data on sociodemography, comorbidity, interventions and medications were also collected.ResultsHigh-risk patients (n=6573, risk score ≥6) had fewer DAOH (∆=-142 days (95% CI: -152 to -131)), and were more likely to readmit or die (all p<0.001) than their low-risk counterparts (n=13 367, risk score <6). Compared with patients who were never prescribed a medication, those who consumed maximal dose of betablockers (∆=39 days (95% CI: 11 to 67)), angiotensin-converting enzyme inhibitors/angiotensin receptor blockers (∆=74 days (95% CI: 49 to 99)) or statins (∆=109 days (95% CI: 90 to 128)) had significantly greater DAOH. Patients who received percutaneous coronary intervention (∆=99 days (95% CI: 81 to 116)) or coronary artery bypass grafting (∆=120 days (95% CI: 92 to 148)) also had significantly greater DAOH than those who did not. The effect sizes of these therapies were significantly greater in high-risk patients, compared with low-risk patients (interaction p<0.001). Analysis of secondary outcomes also found significant interaction between both medical and interventional therapies with readmission and death, implicating greater benefits for high-risk patients.ConclusionsCHD patients can be effectively risk-stratified, and use of this information for a risk-guided strategy to prioritise high-risk patients may maximise benefits from additional resources spent on intensive disease management.

Dataset Information

Extracting disease risk profiles from expression data for linkage analysis: application to prostate cancer.

Publications

Extracting disease risk profiles from expression data for linkage analysis: application to prostate cancer.

Similar Datasets

OmicsDI is part of the ELIXIR infrastructure

Tweets