Clust_100_Ge_Datasets: 100 Gene Expression Datasets from Five Model Species

100 microarray and RNA-seq gene expression datasets from five model species (human, mouse, fruit fly, arabidopsis, and baker's yeast) form a benchmark set. The resource includes raw data, pre-processed files, clustering results, evaluations, and scripts from a University of Oxford study. It also contains results for 19 additional 'large' datasets and simultaneous multi-dataset analyses.

Use Cases

• Benchmarking clustering algorithms based on the provided results for seven widely used methods.
• Evaluating gene ontology (GO) and REACTOME pathway enrichment based on the included analysis files.
• Testing parameter sensitivity for clustering methods using the 'ParamSweepClust' sub-folders.
• Analyzing clustering performance on larger datasets using the 19 'large' dataset results.

Strengths

• Includes 100 benchmark datasets from five model species, providing a broad biological scope.
• Contains raw data, processed data, clustering results, and evaluation metrics for each dataset.
• Extends analysis to 19 additional 'large' datasets and includes results for simultaneous multi-dataset clustering.

Limitations

• Row count and column-level documentation are unknown, requiring manual inspection after download.
• Last update date is unknown; freshness unverified.

Provenance

Source

University of Oxford

Collection Method

Benchmark set compiled for testing the 'clust' clustering method.