What is AMiGA?
It is a Python-based program for high-throughput analysis of microbial growth data sets that are typically generated by multi-well plate readers. It can estimate classic microbial growth parameters such as exponential growth time, doubling time, lag phase, and carrying capacity. It can also estimate microbial death, death rates, and adaptation time, as well as detect and describe diauxic shifts in growth data. It is especially useful for the rapid analysis of standardized multi-well plates including Biolog Phenotypic Microarray (PM) data.
An advantage of AMiGA is its application of a nonparameteric approch for modelling microbial growth data. Many microbial growth curves do not follow standard logistic or sigmoidal shapes. The nonparameteric approach of Gaussian Process (GP) regression can model microbial growth data and estimate growth parameters without assumptions about the shape of their growth curves. AMiGA streamlines the analysis of microbial growth data using Gaussian Processes in a user-friendly fashion and makes it accessible to researchers with little or no background in bioinformatics. In order to plot or analyze the data, users simply interact with AMiGA using the command-line terminal by pointing to a file of interest. User can pass additional arguments or files or define program-specific parameters for more sophisticated analysis.
What are the main functions of AMiGA?
| Function |
Description |
|---|---|
Summarize |
Plot the growth curves for 96-well plates and summarize growth curves with basic metrics such as maximum optical density or fold-change relative to control well(s). |
Fit |
Infer growth kinetic parameters such as adaptation time, lag time, exponential growth rate, doubling time, death rate, carrying capacity, and area under the curve. Inference can be performed on individual growth curves or pooled replicates. Pooling can provide a summary distribution for each growth parameter: mean and standard deviation or confidence interval. This function also automatically detects multiple phases of growth (e.g. diauxic shifts) and infers the growth parameters for each unique growth phase. |
Normalize |
Adjust growth parameters of treatment curves based on growth parameters of control curves. |
Compare |
Provide a statistical summary of the differences in growth parameters between two growth curves. |
Test |
Perform Bayesian hypothesis-driven statistical testing of differences in microbial growth under distinct experimental conditions. |
Heatmap |
Plot a heatmap to help you visualize differences in a growth parameter across distinct conditions. |
Get-Confidence |
Estimate confidence intervals for growth parameters or growth curves. |
Get-Time |
Calculate time needed to reach a certain optical density. |
Get Started
Start by installing AMiGA or take a look at an example.
Documentation
- Install AMiGA
- Prepare Data: Preparing and organizing your data
- Prepare Metadata: Preparing meta-data for your files
- Summarize & Plot: Plotting and summarizing basic metrics about growth curves
- Fit Curves: Fitting growth curves with GP regression
- Batch Analyze: Fitting growth curves from multiple files at the same time
- Subset Data: Subsetting data sets to analyze specific growth curves
- Detect Diauxie: Detecting diauxic shifts or multi-phasic growth
- Pool Replicates: Pooling replicate growth curves then fitting
- Normalize Parameters: Normalizing growth parameters
- Compare Parameters: Testing for significant differences between growth parameters
- Estimate Confidence: Estimating confidence intervals for growth parameters or growth curves
- Get Time at Threshold: Estimating time needed to reach certain OD threshold
- Test Hypotheses: Testing for functional differences between two growth curves
- Plot Heatmaps: Comparing growth parameters of different conditions with heatmaps
- Command-line interface: arguments and how to use them?
- Code Vignette: Sneak peak into the code and how to implement in your pipeline
- Configure default parameters: Adjusting the default parameters stored in
libs/config.py - Example Pipeline: Tutorial on analyzing growth curves for Examploides randomii
Report Bugs or Request Features
Please post issues or requests here. This is a good place to post any problems with installing or running AMiGA. If you are new to installing and running Python-based software, you may run into errors that are common but easily resolved, so do not hesitate to post any issues. Otherwise, if you have any general questions about using AMiGA, please reach out to me via e-mail.
Citation
FS Midani, J Collins, RA Britton. AMiGA: software for automated Analysis of Microbial Growth Assays. mSystems (2020). https://doi.org/10.1128/mSystems.00508-21