Contents Menu Expand Light mode Dark mode Auto light/dark, in light mode Auto light/dark, in dark mode Skip to content
sweepLink documentation
Light Logo Dark Logo

Getting Started

  • Installation
  • Tutorial
    • Step 1: Preparing the Input
    • Step 2: Inference
    • Step 3: Results
    • Step 4: Refining
  • Using your own data

User Manual

  • How it works
  • sweepLink tasks
    • Inference
    • Simulation
    • Likelihood surfaces
  • Companion scripts
    • sweeplink-vcf2input
    • sweeplink-extract
    • sweeplink-plot
    • sweeplink-plot-posteriors
  • File formats
  • FAQ
Back to top

sweepLink tasks¶

If you are starting out, work through the tutorial first — this section is for looking things up.

SweepLink performs several tasks. The task is the first argument on the command line, and each one accepts its own options:

$ ./sweeplink <task> --option value ...

Task

What it does

infer

Infer locus-specific selection coefficients and population size from time-series data. This is the main task.

simulate

Simulate time-series allele count data under the model.

calculateLL

Evaluate the log-likelihood of each locus for a given set of parameters, without running an MCMC.

calculateLLForS

Evaluate the log-likelihood of each locus at every value on the grid of selection coefficients.

calculateLLForSH

As above, over every combination of selection and dominance.

calculateLLForN

Evaluate the log-likelihood over a grid of population sizes, per locus or summed across loci.

The four calculateLL* tasks are for inspecting the likelihood surface directly rather than sampling from the posterior; they are an advanced tool and most analyses never need them.

Instead of flags, any task also accepts a .parameters file — the complete, re-runnable record every run writes:

$ ./sweeplink myrun.parameters

Everything around the inference — converting a VCF, extracting results, plotting — lives in companion scripts instead.

Next
Inference
Previous
How it works
Copyright © 2024, Ekaterina Noskova
Made with Sphinx and @pradyunsg's Furo