Curation of over 10 000 transcriptomic studies to enable data reuse

Area of Science:

• Bioinformatics
• Genomics
• Computational Biology

Background:

• Public transcriptomic repositories contain vast data, but reuse is hindered by inconsistent metadata, processing, and probe-gene mappings.
• Manual curation and reprocessing are essential for effective transcriptomic data utilization.
• The Gemma system was developed to overcome these challenges in transcriptomic data accessibility.

Purpose of the Study:

• To provide an updated overview of the Gemma bioinformatics system's resources and capabilities.
• To detail Gemma's holdings, data processing pipelines, curation guidelines, and software features.
• To highlight Gemma's role in facilitating the reuse of transcriptomic data.

Main Methods:

• Gemma utilizes a curated database storing transcriptomic datasets from human, mouse, and rat.
• Data processing and analysis pipelines are employed for quality control and standardization.
• Ontologies are used for topic annotation, enabling structured data representation.

Main Results:

• As of June 2020, Gemma contains 10,811 manually curated datasets and over 395,000 samples.
• The system supports hundreds of transcriptomic platforms, including microarray and RNA sequencing.
• A significant portion (34%) of Gemma's holdings are brain-related datasets, with broad coverage of conditions and tissues.

Conclusions:

• Gemma offers a valuable resource for accessing and analyzing curated transcriptomic data.
• The system's comprehensive curation and analysis pipelines improve data usability and research reproducibility.
• Users can access Gemma's data and analyses via its website, RESTful service, and R package.