varan

varan computes the Expected Information Value for variant sites from samtools' mpileup to assess and prioritize genetic variation in high-throughput sequencing data.

Key Features:

• ESIV (Expected Information Value): Calculates ESIV by integrating uncertainty measures such as p-values with effect-size metrics like Fst into a single score for variant site informativeness.
• Handling Uncertainty: Implements a conservative estimator that accounts for sampling bias and sequencing errors to reduce false-positive variant calls.
• mpileup Processing: Processes sequence data in samtools' mpileup format and can read mpileup data directly from a pipe to avoid generating large temporary files.
• Site-level Statistics: Computes site-level statistics relevant to genetic diversity to support assessment of variant frequency and informativeness.

Scientific Applications:

• Ecology: Identifies and prioritizes genetic variants across populations in ecological studies.
• Population genetics: Distinguishes true genetic variation from sequencing noise to inform population-genetic analyses.
• Candidate site selection: Enables selection of candidate diagnostic or informative sites for downstream analyses by separating true variants from artifacts.

Methodology:

Reads samtools' mpileup input (including via a pipe), computes the expected information gained per site (ESIV) by integrating p-values and Fst, applies a conservative estimator to account for sampling bias and sequencing errors, and has been applied to both simulated and real datasets.