Software Information

GET STARTED
1
Request Info
2
Visit
3
Apply
GET STARTED
1
Request Info
2
Visit
3
Apply

BAMarray

Java software for microarray data. Implements Bayesian Analysis of Variance for Microarrays (BAM)

Download »

Joint Mean-Variance Stabilization and Regularization for High Throughput Data

New R package!

An R package that stabilizes variances and also regularizes inferences using information in both the mean and the variance. Applications include high throughput “omic” data as well as other high throughput datasets.

Download »

LDLasso

An R package that performs a linkage-disequilibrium based fusion of SNPestimates in GWAS studies with the goal of producing flatter, longer signal sites to allow weaker, correlated signal while still inducing sparsity via lasso shrinkage.

Download »

Local Sparse Bump Hunting

An R package that searches for local structures in the form of modes in high dimensional space. Sparsity is imposed via sparse principal components analysis and localness imposed by tree-based partitioning of the predictor space. Modes are then identified via a version of the patient rule induction method (PRIM). New methods for optimal PRIM hyperparameter tuning are also implemented.

Download »

pathwayPCA

With the advance in high-throughput technology for molecular assays, multi-omics datasets have become increasingly available. However, most currently available pathway analysis software do not provide estimates on sample-specific pathway activities, and provide little or no functionalities for analyzing multiple types of omics data simultaneously. To address these challenges, we present pathwayPCA, a unique integrative pathway analysis software that utilizes modern statistical methodology on principal component analysis (PCA) and gene selection.

The main features of pathwayPCA include:

  1. Performing pathway analysis for datasets with binary, continuous, or survival outcomes with computational efficiency.
  2. Extracting relevant genes from pathways using the SuperPCA and AESPCA approaches.
  3. Computing PCs based on the selected genes. These estimated latent variables represent pathway activity for individual subjects, which can be used to perform integrative pathway analysis, such as multi-omics analysis, or predicting survival time.
  4. Can be used to analyze studies with complex experimental designs that include multiple covariates and/or interaction effects. For example, testing whether pathway associations with clinical phenotype are different between male and female subjects.
  5. Performing analyses with enhanced computational efficiency with parallel computing and enhanced data safety with S4-class data objects.

Download »

randomForestSRC

Random forest for survival, regression, and classification. A unified treatment of Breiman’s random forests.

Download »

spikeslab

Spike and slab R package for high dimensional linear regression models (release 1.1.2). Uses a generalized elastic net (gnet) for variableselection. Parallel processing enabled.

Download »