subSeq

subSeq performs subsampling of high-throughput sequencing count data (e.g., RNA-Seq) as an R package to assess required read depth and experimental saturation for statistical power and accuracy.

Key Features:

• Subsampling of Sequencing Data: Performs random subsampling and simulation of lower read depths from sequencing count data to emulate reduced sequencing depths.
• Optimal Read Depth Determination: Identifies saturation points where additional sequencing yields diminishing returns, informing the read depth that balances cost with power and accuracy.
• Verification of Existing Experiments: Evaluates completed experiments by assessing whether existing sequencing depth was sufficient to support experimental conclusions.
• Efficient Methodology: Implements an efficient subsampling approach that calculates informative metrics at each sampled depth.

Scientific Applications:

• Experiment Design: Guides design of RNA-Seq and other NGS experiments by determining cost-effective sequencing depths that achieve sufficient statistical power.
• Data Analysis: Enables retrospective assessment of existing datasets to verify adequacy of sequencing depth and robustness of results.

Methodology:

Random subsampling (simulation of lower read depths) from sequencing count data at various depths and calculation of metrics on each subsample to identify saturation points where additional sequencing no longer substantially improves power or accuracy.