ctsGE

ctsGE performs clustering of gene expression time-series data as an R package, identifying temporal expression patterns while avoiding conventional filtering to preserve informative variation.

Key Features:

• Implementation: Implemented as an R package for analysis of time-series gene-expression datasets.
• Sorting step: Divides the dataset into smaller groups based on each gene's relationship to the median of the time series rather than filtering out noisy profiles.
• Expression index definition: Represents each gene's temporal behavior as a sequence of 1, -1, and 0 values reflecting changes relative to the median over time.
• Grouping by expression index: Groups genes that share identical expression index sequences to collect similar temporal patterns.
• K-means clustering: Applies k-means within each group of identical indices to further subdivide genes into more granular subclusters.
• Two-step clustering process: Combines index-based grouping and within-group k-means subdivision to produce final clusters of temporal expression patterns.

Scientific Applications:

• Time-series genomics: Organizing and analyzing temporal gene-expression data to identify dynamic expression patterns across time points.
• Biological process elucidation: Characterizing temporal changes in gene expression to support studies of biological processes and regulatory dynamics.
• Disease mechanism investigation: Investigating temporal expression changes relevant to disease mechanisms and progression.

Methodology:

Sort genes by relation to the median; define an expression index per gene as a sequence of 1, -1, and 0 relative to the median; group genes with identical indices; and apply k-means clustering within each group.