Dataset Information

Promoting Similarity of Sparsity Structures in Integrative Analysis with Penalization.

ABSTRACT: For data with high-dimensional covariates but small sample sizes, the analysis of single datasets often generates unsatisfactory results. The integrative analysis of multiple independent datasets provides an effective way of pooling information and outperforms single-dataset and several alternative multi-datasets methods. Under many scenarios, multiple datasets are expected to share common important covariates, that is, the corresponding models have similarity in their sparsity structures. However, the existing methods do not have a mechanism to promote the similarity in sparsity structures in integrative analysis. In this study, we consider penalized variable selection and estimation in integrative analysis. We develop an L0-penalty based method, which explicitly promotes the similarity in sparsity structures. Computationally it is realized using a coordinate descent algorithm. Theoretically it has the selection and estimation consistency properties. Under a wide spectrum of simulation scenarios, it has identification and estimation performance comparable to or better than the alternatives. In the analysis of three lung cancer datasets with gene expression measurements, it identifies genes with sound biological implications and satisfactory prediction performance.

SUBMITTER: Huang Y

PROVIDER: S-EPMC6086364 | biostudies-literature | 2017

REPOSITORIES: biostudies-literature

ACCESS DATA

Publications

Promoting Similarity of Sparsity Structures in Integrative Analysis with Penalization.

Huang Yuan Y Zhang Qingzhao Q Zhang Sanguo S Huang Jian J Ma Shuangge S

Journal of the American Statistical Association 20170503 517

For data with high-dimensional covariates but small sample sizes, the analysis of single datasets often generates unsatisfactory results. The integrative analysis of multiple independent datasets provides an effective way of pooling information and outperforms single-dataset and several alternative multi-datasets methods. Under many scenarios, multiple datasets are expected to share common important covariates, that is, the corresponding models have similarity in their sparsity structures. Howev ...[more]

PMID: 30100648

Similar Datasets

Project description:BackgroundAmblyopia is defined as defective visual acuity in one or both eyes without demonstrable abnormality of the visual pathway, and is not immediately resolved by wearing glasses.ObjectivesTo assess the effectiveness and safety of conventional occlusion versus atropine penalization for amblyopia.Search strategyWe searched CENTRAL, MEDLINE, EMBASE, LILACS, the WHO International Clinical Trials Registry Platform, preference lists, science citation index and ongoing trials up to June 2009.Selection criteriaWe included randomized/quasi-randomized controlled trials comparing conventional occlusion to atropine penalization for amblyopia.Data collection and analysisTwo authors independently screened abstracts and full text articles, abstracted data, and assessed the risk of bias.Main resultsThree trials with a total of 525 amblyopic eyes were included. One trial was assessed as having a low risk of bias among these three trials, and one was assessed as having a high risk of bias.Evidence from three trials suggests atropine penalization is as effective as conventional occlusion. One trial found similar improvement in vision at six and 24 months. At six months, visual acuity in the amblyopic eye improved from baseline 3.16 lines in the occlusion and 2.84 lines in the atropine group (mean difference 0.034 logMAR; 95% confidence interval (CI) 0.005 to 0.064 logMAR). At 24 months, additional improvement was seen in both groups; but there continued to be no meaningful difference (mean difference 0.01 logMAR; 95% CI -0.02 to 0.04 logMAR). The second trial reported atropine to be more effective than occlusion. At six months, visual acuity improved 1.8 lines in the patching group and 3.4 lines in the atropine penalization group, and was in favor of atropine (mean difference -0.16 logMAR; 95% CI -0.23 to -0.09 logMAR). Different occlusion modalities were used in these two trials. The third trial had inherent methodological flaws and limited inference could be drawn.No difference in ocular alignment, stereo acuity and sound eye visual acuity between occlusion and atropine penalization was found. Although both treatments were well tolerated, compliance was better in atropine. Atropine penalization costs less than conventional occlusion. The results indicate that atropine penalization is as effective as conventional occlusion.Authors' conclusionsBoth conventional occlusion and atropine penalization produce visual acuity improvement in the amblyopic eye. Atropine penalization appears to be as effective as conventional occlusion, although the magnitude of improvement differed among the three trials. Atropine penalization can be used as first line treatment for amblyopia.

Dataset Information

Promoting Similarity of Sparsity Structures in Integrative Analysis with Penalization.

Publications

Promoting Similarity of Sparsity Structures in Integrative Analysis with Penalization.

Similar Datasets

OmicsDI is part of the ELIXIR infrastructure

Tweets