preseq

preseq predicts the number of distinct reads and expected genome coverage obtainable from additional sequencing using initial shallow sequencing data to guide allocation of sequencing depth.

Key Features:

• Predictive modeling: Uses a non-parametric empirical Bayes Poisson model to estimate coverage gains and the number of distinct reads from deeper sequencing.
• Shallow-read analysis: Derives initial coverage and complexity statistics from shallow sequencing reads mapped to a reference genome.
• Sequencing-depth optimization: Estimates the marginal utility of additional sequencing to support cost-effective selection of sequencing depth.
• Library screening and comparison: Compares multiple libraries to identify high-complexity samples and flag low-complexity libraries unlikely to yield substantial new information upon deeper sequencing.

Scientific Applications:

• Genomics: Forecasts coverage statistics to plan sequencing strategies in genomics research.
• Single-cell DNA sequencing: Predicts genome coverage from early data where limited starting material and library-prep variability affect outcomes.
• Detection of genetic variation: Informs experiment design for detecting genetic variation that may be difficult to resolve with bulk sequencing alone.
• Library and protocol evaluation: Enables systematic comparison of sequencing libraries and strategies to decide whether additional sequencing is likely to be beneficial.

Methodology:

Analyzes shallow sequencing reads mapped to a reference genome to derive initial coverage and complexity statistics, then applies a non-parametric empirical Bayes Poisson framework to extrapolate expected distinct reads and genome coverage at increased sequencing depths.