HiMaLAYAS

Hierarchical Matrix Layout and Annotation Software (HiMaLAYAS) is a framework for post hoc enrichment-based annotation and visualization of hierarchically clustered matrices. HiMaLAYAS treats dendrogram-defined clusters as statistical units, tests categorical annotations for enrichment, controls multiple testing, and renders significant annotations alongside clusters. HiMaLAYAS supports both biological and non-biological domains.

For a full description of HiMaLAYAS and its applications, see:
Horecka, I., and Röst, H. (2026)
HiMaLAYAS: enrichment-based annotation and visualization of hierarchically clustered matrices
bioRxiv. https://www.biorxiv.org/content/10.64898/2026.02.11.705303v2
Submitted to Bioinformatics Advances.

Documentation and Tutorial

• Full Documentation: himalayas-base.github.io/himalayas-docs
• Figure Gallery: himalayas-base.github.io/himalayas-docs/11_figure_gallery
• Try in Browser (Binder):
  • Quickstart:
  • Advanced Quickstart:
• Documentation Repository: github.com/himalayas-base/himalayas-docs

Installation

HiMaLAYAS is compatible with Python 3.8 or later and runs on major operating systems.

pip install himalayas
# Optional: faster clustering + richer compressed-label text processing
pip install "himalayas[speed,text]"

Detailed installation options and fallback behavior are documented at himalayas-base.github.io/himalayas-docs/1_installation.

Key Features of HiMaLAYAS

• Real-Valued Matrix Input: Operates on real-valued matrices encoding relationships among observations.
• Dendrogram-Defined Clusters: Cuts the dendrogram at a user-defined threshold to define dendrogram-defined clusters for downstream analysis.
• Overrepresentation Testing: Uses a one-sided hypergeometric test to evaluate term enrichment in each cluster against the matrix background; genes with multiple term annotations are tested independently for each term, with no gene-level exclusivity.
• Multiple-Testing Control: Supports Benjamini-Hochberg false discovery rate (FDR) correction for cluster-term tests, applied globally by default (fdr_scope="global").
• Annotation Mapping and Rendering: Maps significant annotations onto the clustered matrix and supports publication-ready matrix visualizations.
• No Redundancy Filtering: Does not perform Gene Ontology graph-aware redundancy filtering, such as collapsing parent-child terms or grouping terms by semantic similarity; all annotation terms passing the minimum-size and minimum-overlap thresholds are tested and reported independently.

Example Usage

We applied HiMaLAYAS to a hierarchically clustered Saccharomyces cerevisiae genetic interaction profile similarity matrix (Costanzo et al., 2016), focusing on genes with high profile variance. Dendrogram-defined clusters were tested for Gene Ontology Biological Process (GO BP; Ashburner et al., 2000) enrichment, with significant annotations rendered alongside clusters.

HiMaLAYAS workflow and application to a hierarchically clustered yeast genetic interaction profile similarity matrix (Costanzo et al., 2016). A real-valued matrix and categorical annotations serve as inputs. HiMaLAYAS cuts the dendrogram at a user-defined threshold, tests categorical annotations for enrichment, controls multiple testing, and renders significant annotations alongside clusters.

Citation

Primary citation

Horecka, I., and Röst, H. (2026)
HiMaLAYAS: enrichment-based annotation and visualization of hierarchically clustered matrices
bioRxiv. https://www.biorxiv.org/content/10.64898/2026.02.11.705303v2
Submitted to Bioinformatics Advances.

Software archive

HiMaLAYAS software for the bioRxiv preprint.
Zenodo. https://doi.org/10.5281/zenodo.19338505

Contributing

We welcome contributions from the community:

• Issues Tracker
• Source Code

Support

If you encounter issues or have suggestions for new features, please use the Issues Tracker on GitHub.

License

This project is distributed under the BSD 3-Clause License.