Gene network-based cancer prognosis analysis with sparse boosting

Ma, Shuangge; Huang, Yuan; Huang, Jian; Fang, Kuangnan

doi:10.1017/s0016672312000419

Cited by 16 publications

(12 citation statements)

References 31 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…In the literature, a growing number of methods have been reported for network-based survival analysis [31][32][33][34][35][36]. Some of them have been implemented as tools, including Net-Cox [31], Reactome FI [32], HyperModules [36] and HotNet [35].…”

Section: Resultsmentioning

confidence: 99%

The `dnet¿ approach promotes emerging research on cancer patient survival

Fang¹,

Gough²

2014

Genome Medicine

View full text Add to dashboard Cite

We present the 'dnet' package and apply it to the 'TCGA' mutation and clinical data of >3,000 patients. We uncover the existence of an underlying gene network that at least partially controls cancer 'survivalness', with mutations that are significantly correlated with patient survival, yet independent of tumour origin and type. The survivalness network has natural community structure corresponding to tumour hallmarks, and contains genes that are potentially druggable in the clinic. This network has evolutionary roots in Deuterostomia identifying PTK2 and VAV1 as under-valued relative to more studied genes from that era. The 'dnet' R package is available at http://cran.r-project.org/package=dnet.

show abstract

Section: Resultsmentioning

confidence: 99%

The `dnet¿ approach promotes emerging research on cancer patient survival

Fang¹,

Gough²

2014

Genome Medicine

View full text Add to dashboard Cite

show abstract

“…We adopt the duo

(selection, stopping) = (BIC + pe n_{s}, HDBIC with or without pe n_{s}),

where the BIC criterion is log( R S S ) + d f × log( n )/ n , and the HDBIC criterion is log( R S S ) + d f × log( n )log( d )/ n []. Adopting the BIC criterion for selecting weak learners has been motivated by published studies []. The HDBIC criterion imposes more penalty than BIC and can generate sparser models.…”

Section: Integrative Analysis and Marker Selection Using Sparse Boostingmentioning

confidence: 99%

Promoting similarity of model sparsity structures in integrative analysis of cancer genetic data

Huang

Liu

et al. 2016

Statistics in Medicine

Self Cite

View full text Add to dashboard Cite

In profiling studies, the analysis of a single dataset often leads to unsatisfactory results because of the small sample size. Multi-dataset analysis utilizes information of multiple independent datasets and outperforms single-dataset analysis. Among the available multi-dataset analysis methods, integrative analysis methods aggregate and analyze raw data and outperform meta-analysis methods, which analyze multiple datasets separately and then pool summary statistics. In this study, we conduct integrative analysis and marker selection under the heterogeneity structure, which allows different datasets to have overlapping but not necessarily identical sets of markers. Under certain scenarios, it is reasonable to expect some similarity of identified marker sets – or equivalently, similarity of model sparsity structures – across multiple datasets. However, the existing methods do not have a mechanism to explicitly promote such similarity. To tackle this problem, we develop a sparse boosting method. This method uses a BIC/HDBIC criterion to select weak learners in boosting and encourages sparsity. A new penalty is introduced to promote the similarity of model sparsity structures across datasets. The proposed method has an intuitive formulation and is broadly applicable and computationally affordable. In numerical studies, we analyze right censored survival data under the AFT (accelerated failure time) model. Simulation shows that the proposed method outperforms alternative boosting and penalization methods with more accurate marker identification. The analysis of three breast cancer prognosis datasets shows that the proposed method can identify marker sets with increased similarity across datasets and improved prediction performance.

show abstract

“…In this study, we focus on formulation (6). In network analysis of gene expression data, it has been suggested that genes with higher connectivity tend to have more important biological implications [Zhang and Horvath 2005; Ma et al 2012]. It is therefore sensible to consider the unnormalized Laplacian.…”

Section: Sparse Group Laplacian Shrinkagementioning

confidence: 99%

“…Huang et al [2011] proposed a sparse Laplacian shrinkage method for variable selection and estimation. A two-step sparse boosting approach was developed in Ma et al [2012]. Bayesian approaches have also been developed [Edwards et al 2012].…”

Section: Introductionmentioning

confidence: 99%

Incorporating Network Structure in Integrative Analysis of Cancer Prognosis Data

Liu

Huang

2012

Genetic Epidemiology

Self Cite

View full text Add to dashboard Cite

In high-throughput cancer genomic studies, markers identified from the analysis of single datasets may have unsatisfactory properties because of low sample sizes. Integrative analysis pools and analyzes raw data from multiple studies, and can effectively increase sample size and lead to improved marker identification results. In this study, we consider the integrative analysis of multiple high-throughput cancer prognosis studies. In the existing integrative analysis studies, the interplay among genes, which can be described using the network structure, has not been effectively accounted for. In network analysis, tightly-connected nodes (genes) are more likely to have related biological functions and similar regression coefficients. The goal of this study is to develop an analysis approach that can incorporate the gene network structure in integrative analysis. To this end, we adopt an AFT (accelerated failure time) model to describe survival. A weighted least squares approach, which has low computational cost, is adopted for estimation. For marker selection, we propose a new penalization approach. The proposed penalty is composed of two parts. The first part is a group MCP penalty, and conducts gene selection. The second part is a Laplacian penalty, and smoothes the differences of coefficients for tightly-connected genes. A group coordinate descent approach is developed to compute the proposed estimate. Simulation study shows satisfactory performance of the proposed approach when there exist moderate to strong correlations among genes. We analyze three lung cancer prognosis datasets, and demonstrate that incorporating the network structure can lead to the identification of important genes and improved prediction performance.

show abstract

Gene network-based cancer prognosis analysis with sparse boosting

Cited by 16 publications

References 31 publications

The `dnet¿ approach promotes emerging research on cancer patient survival

The `dnet¿ approach promotes emerging research on cancer patient survival

Promoting similarity of model sparsity structures in integrative analysis of cancer genetic data

Incorporating Network Structure in Integrative Analysis of Cancer Prognosis Data

Contact Info

Product

Resources

About