Designing fitness assays

About

Code corresponding to the following paper:

Resolving deleterious and near-neutral effects requires different pooled fitness assay designs

Anurag Limdi and Michael Baym

In this project, we explored how assay parameters impact measurements of near-neutral and deleterious fitness effects, and whether the best design parameters differ for the specific fitness effect sizes that are relevant for the biological question under investigation.

Organization

1. Processed data

This folder is empty: please download the data from https://doi.org/10.5281/zenodo.6547536. This processed data is generated using the scripts in https://github.com/baymlab/2022_Limdi-TnSeq-LTEE (Part 1: Data to Trajectories), and is required for final figure generation and analysis.

2. Metadata

This folder contains the relevant metadata for analysis, including gene names, locations, reference genomes, etc.

3. Analysis notebooks

Contains the following notebooks:

• fitness_assay_simulations: In this notebook, I estimate theoretical error bounds, and simulate fitness assays using several simplifying assumptions, and explore how modifying experimental parameters impacts errors in measurements.
• reanalysis_experimental_data: In this notebook, I reanalyse our previously published and very deeply sequenced E. coli TnSeq dataset, by downsampling and using a subset of timepoints to explore how changing those parameters impacts measurement errors, and compare results against theory/simulations