MultiDST 0.1.11

A toolkit for multiple hypothesis testing that implements common FWER and FDR control procedures, with intuitive significance visualisation using the Significant Index Plot.

MultiDST Package

Multiple Testing, Made Easy

MultiDST is a Python library for large-scale multiple hypothesis testing. It provides a unified and user-friendly framework for controlling error rates when performing many simultaneous statistical tests. The package supports both family-wise error rate (FWER) and false discovery rate (FDR) control, making it suitable for applications such as genomics, biomedical studies, and other high-dimensional analyses. A key feature of MultiDST is the Significant Index Plot (SIP), which enables clear and intuitive visualisation of significant hypotheses under different correction methods.

Implemented Methods

Bonferroni Correction
    Reason for Selection: Baseline method; simple and conservative.
    Error Type: Controls the family-wise error rate (FWER).
    Reference: Bonferroni, C. (1935). Il calcolo delle assicurazioni su gruppi di teste. https://api.semanticscholar.org/CorpusID:89994272

Holm-Bonferroni Correction
    Reason for Selection: Sequential approach; adjusts thresholds progressively.
    Error Type: Controls the family-wise error rate (FWER).
    Reference: Holm, S. (1979). A Simple Sequentially Rejective Multiple Test Procedure. Source: Scandinavian Journal of Statistics Scand J Statist, 6(6), 65–70. http://www.jstor.org/stable/4615733%0Ahttp://www.jstor.org/page/info/about/policies/terms.jsp%0Ahttp://www.jstor.org

Benjamini-Hochberg Procedure
    Reason for Selection: Balances power and control; widely used in genomics.
    Error Type: Controls the false discovery rate (FDR).
    Reference: Benjamini, Y., & Hochberg, Y. (1995). "Controlling the false discovery rate: A practical and powerful approach to multiple testing." Journal of the Royal Statistical Society. Series B (Methodological), 57(1), 289-300.

Benjamini-Yekutieli Method
    Reason for Selection: FDR control under dependence; accounts for correlated tests.
    Error Type: Controls the false discovery rate (FDR).
    Reference: Benjamini, Y., & Yekutieli, D. (2001a). The control of the false discovery rate in multiple testing under dependency. Annals of Statistics, 29(4), 1165–1188. https://doi.org/10.1214/aos/1013699998

Storey's Q Value
    Reason for Selection: Adaptive FDR control; estimates proportion of true null hypotheses.
    Error Type: Controls the false discovery rate (FDR).
    Reference: Storey, J. D., & Tibshirani, R. (2003). Statistical Significance for Genome-Wide Studies.

SGoF Test (Sequential Goodness-of-Fit)
    Reason for Selection: Increased power for large-scale testing; combines p-values from multiple tests.
    Error Type: Varies (depends on setup).
    Reference: Carvajal-Rodríguez, A., de Uña-Alvarez, J., & Rolán-Alvarez, E. (2009). A new multitest correction (SGoF) that increases its statistical power when increasing the number of tests. BMC Bioinformatics, 10, 209. https://doi.org/10.1186/1471-2105-10-209

Installation

You can install the MultiDST Package using pip:

pip install MultiDST_package

Special Features

Multiple Methods: Implements several established methods for multiple hypothesis testing, including Bonferroni correction, Holm-Bonferroni correction, Benjamini-Hochberg procedure, and more.

Flexible Usage: Designed to handle different types of data and scenarios typically encountered in statistical analysis.

Integration: Methods are integrated into a unified framework, making it easier to apply and compare different approaches within the same analysis.

Enhancements: Includes novel enhancements such as a "hybrid" procedure and "multi-weighting" to improve robustness and flexibility in hypothesis testing.

Visualization: Provides tools for visualizing significant hypotheses, aiding in the interpretation of results.