t-Distributed Stochastic Neighbor Embedding (t-SNE): A tool for eco-physiological transcriptomic analysis.

Area of Science:

• Marine biology
• Bioinformatics
• Genomics

Background:

• High-throughput RNA sequencing (RNA-Seq) generates vast amounts of complex data.
• Analyzing RNA-Seq data for plankton ecophysiology requires advanced bioinformatics workflows.
• Existing methods for interpreting transcriptomic data can be challenging.

Purpose of the Study:

• To explore the application of t-distributed Stochastic Neighbor Embedding (t-SNE) for analyzing marine plankton RNA-Seq data.
• To demonstrate how t-SNE can simplify the interpretation of complex transcriptomic datasets.
• To evaluate t-SNE's utility in comparing gene expression profiles across different developmental stages, experimental conditions, and environmental samples.

Main Methods:

• Applied t-distributed Stochastic Neighbor Embedding (t-SNE) to existing RNA-Seq datasets from copepods (Calanus finmarchicus and Neocalanus flemingeri).
• Compared gene expression profiles across developmental stages, experimental treatments, and environmental locations.
• Validated t-SNE identified profile categories using differential gene expression and Gene Ontology (GO) analyses.

Main Results:

• t-SNE effectively reduced complexity and identified distinct categories within RNA-Seq data.
• The method allowed for evaluation of global gene expression differences and specific biological processes (e.g., lipid metabolism, stress response).
• t-SNE classifications were consistent with results from differential gene expression and GO analyses.

Conclusions:

• t-distributed Stochastic Neighbor Embedding (t-SNE) is a powerful tool for analyzing and interpreting complex RNA-Seq data from marine plankton.
• This approach facilitates the classification of samples based on transcriptional physiology, independent of sampling context.
• t-SNE aids in identifying trends and understanding biological responses within plankton communities and species.