Cluster Analysis with Resampling for Validation and Exploration (CARVE)

2026-05-29

AI Summary

Clustering outcomes are highly sensitive to algorithmic choices, preprocessing steps, and the number of clusters, yet conventional validation metrics often fail in high-dimensional, heavy-tailed, or nonlinear biomedical data, leading to irreproducible findings. This work proposes a resampling-driven framework for clustering evaluation that unifies stability and generalization analyses for the first time, enabling diagnostic assessments at global, cluster-level, and sample-level resolutions while producing consensus cluster labels and selection criteria. By circumventing restrictive geometric assumptions inherent in traditional methods, the approach offers a scikit-learn–compatible Python API and a Seurat-compatible R interface. It consistently approximates optimal clustering across six synthetic benchmarks, significantly outperforming existing metrics, and uncovers finer biological structures in real-world genomics and proteomics datasets.

Abstract

Clustering is widely used across the sciences as the foundation for downstream data-driven scientific discoveries. However, clustering results are highly sensitive to the choice of algorithm, preprocessing, and the number of clusters $k$, producing scientific claims that are often not reproducible. The current state of the art for validating clustering solutions consists of clustering validation indices (CVIs) such as Silhouette, Davies-Bouldin, and Calinski-Harabasz, which rely on geometric assumptions that break down on the heavy-tailed, high-dimensional, and nonlinearly structured data encountered in biomedical research. Resampling-based alternatives - grounded in the ideas of clustering stability and generalizability - have been proposed but remain scattered across specialized tools with no unified, accessible software. We fill this gap with CARVE (Cluster Analysis with Resampling for Validation and Exploration), an open-source Python and R package that jointly evaluates multiple clustering algorithms and hyperparameters, returning stability and generalizability diagnostics at the global, cluster, and sample level together with principled selection rules and consensus-based cluster labels. Across six synthetic benchmarks CARVE consistently recovers near-optimal clusterings where classical indices degrade substantially. On experimental genomics and proteomics data sets, CARVE recovers finer biological structure when classical CVIs collapse entirely. CARVE is available with a scikit-learn-compatible Python API and an analogous R interface compatible with Seurat workflows.

Problem

• clustering validation
• resampling
• reproducibility
• high-dimensional data
• cluster stability

Innovation

• resampling
• clustering validation
• stability
• generalizability
• consensus clustering