Software

Hypothesis testing (multivariate, high-dimensional, non-Euclidean data)

Two-sample tests

• zidstats (Python): Two-sample testing, ranking, and distributional diagnostics (GitHub).

• gTests: Graph-based two-sample tests.

• kerTests: Kernel-based two-sample tests.

• GraphRankTest: Rank in similarity graph edge-count two-sample test (RISE).

• gCat: Two-sample tests for categorical data utilizing similarity information among the categories. Useful when the number of categories is large and the contingency table sparsely populated.

Covariate balance in observational studies

• BalanceCheck: Balance check for multiple covariates.

For paired data or paired matching

• gTestsPair: Two-sample tests for paired data and pair matching.

Multi-sample tests

• gTestsMulti: Graph-based multi-sample tests.

Change-point analysis (multivariate, high-dimensional, non-Euclidean data)

Offline change-point detection (segmentation)

• gSeg: Graph-based change-point detection.

• kerSeg: Kernel-based change-point detection.

• ringSeg: Rank in similarity graph edge-count change-point detection.

Online/Sequential detection (for streaming data)

• gStream: Graph-based change-point detection for streaming data.

DNA copy number profiling

Allele-specific copy number profiling

• falconx: for exome sequencing data.

• falcon: for genome sequencing data.

Pipeline (on GitHub)

• Marathon: Integrative pipeline for profiling DNA copy number and inferring tumor phylogeny.