optCluster

optCluster evaluates clustering partitions by integrating statistical and biological validation measures to identify optimal clusterings for continuous and count genomic and RNA sequencing data.

Key Features:

• R package: Implemented as an R package for clustering validation in bioinformatics.
• Integration of validation measures: Integrates multiple statistical and biological validation measures for cluster assessment.
• Data types supported: Tailored for both continuous and count data.
• Partition comparison: Compares multiple clustering partitions generated by different algorithms or varying numbers of clusters.
• Weighted rank aggregation: Applies weighted rank aggregation to combine scores from numerous performance measures into a single aggregated ranking.
• Biological validation for RNA-seq: Includes biological validation measures specifically beneficial for RNA sequencing data.
• Specialized genomic algorithms: Contains specialized algorithms developed for genomic data.
• Single-function comparison: Provides a single function to compare various clustering partitions simultaneously.
• High-throughput sequencing support: Designed to operate on high-throughput sequencing datasets.

Scientific Applications:

• RNA sequencing analysis: Assessing cluster solutions for RNA sequencing (RNA-seq) data using biological and statistical validations.
• Genomic data clustering: Clustering and validation of genomic data using specialized algorithms.
• High-throughput sequencing studies: Analyzing high-throughput sequencing datasets to identify biologically relevant clusters.
• Cluster selection in bioinformatics: Objective selection of optimal clustering partitions in bioinformatics research.

Methodology:

Combines multiple performance scores using weighted rank aggregation and integrates statistical and biological validation measures to rank clustering partitions, supporting comparison of partitions produced by various algorithms and different numbers of clusters.