Iris: expanding the palette of microbial phenotypic readouts

What does Iris do?

Iris is an image processing software that automatically quantifies several features of microbial colonies in high throughput. While most available software can quantify colony size, Iris adds several more phenotypes. Also Iris can be easily expanded to new phenotypes, by writing a new readout module.

How can I download the software?

That's easy, just follow this link to the latest binary distribution (.jar files), as well as the source code: https://github.com/critichu/Iris/releases/latest

Here you can find some example files to play around with (includes example output)

What do I need to run Iris?

Iris runs on Java. You can get the latest version by following this link:
[https://java.com/en/download] (https://java.com/en/download)

Using Iris

Iris is written in Java, which makes it compatible with most operating systems (Windows, macOS, *nix). Java software is typically distributed in JAR files (.jar ending), which you can run by a double-click (see below command-line alternatives).

Next, click "open folder", and select a folder which holds the pictures to be processed. Iris will process any pictures with a ".jpg" file ending, displaying any messages on the application window. You can abort the process at any time, by closing the program.

Running Iris

Iris GUI

To run Iris as a window (GUI) application, open a command line window, cd into the folder the Iris JAR file is in, and run the following command:

java -Xmx2g -jar iris.jar

Note: the Iris jar file can also be run on all systems by double clicking. Iris will attempt to set the heap space automatically. If you run into "out of memory" error messages, please make sure you have permanently set the maximum Java heap size to 2g (2 Gigabyte). Google on how to do that in your particular system/setup.

Iris console (e.g. for scripting)

To run Iris as a console application (no GUI window), which can be rather useful for scripting, you need to provide Iris with a profile name and a folder that pictures are found in:

java -Xmx1g -jar iris.jar "profile name" "folder location”

NOTE: please keep the quotes when inputting the profile name

Valid profile names are the following:

• "Colony growth",
• "Colony growth inverted",
• "B.subtilis sporulation",
• "CPRG profile",
• "Biofilm formation",
• "Morphology profile",
• "Morphology&Color profile",
• "Xgal assay"

Here is a visual overview of those profiles, and the Iris workflow:

Here you can find a table listing all algorithms used in each Iris profile:

Iris output

Iris outputs 2 kind of files and a log file.

.iris files

These hold the results of the feature extraction process. They are in fact plain text tab-delimited files, which can easily be opened with any text viewer or Excel.

.grid files

these are currently saved mainly for debugging purposes, as they demonstrate the results of gridding algorithm. In grid files, colony boundaries are painted in cyan color, so grid files can also be used to check the results of the colony detection algorithms.

.log files

Each time Iris is run, it creates a timestamped log file. This is a text file that is currently used for debugging reasons; any error messages will also be written on this file.

Visualizing Iris output

For this you will need:

• a folder containing iris files (can be only one file)
• the name of a column in the iris file

Download these files and extract the zip. On a terminal, change your directory to the unzipped folder, and run the following:

./plot.colony.phenotype.R --folder [folderName] --phenotype [columnName]

This will make pdf files saved next to your iris files.
Here is an example of what you can do using this script:

---

Iris ColonyPicker

In case your colonies are not in a grid format (96, 384, 1536, or 6144), you can use the Iris' sister software: ColonyPicker. Just double-click on it's jar file, select an iris profile to quantify a colony, and your image file.

How to use it:

• Picking a colony using round (oval) selection will tell Iris to get the features within those manually defined bounds.
• Using a rectangular selection surrounding one colony will get Iris to automatically detect the colony bounds and subsequently analyze this colony.

---

Iris kinetics tool

Using the Iris kinetics tool

The kinetics companion tool is written in R and uses the grofit R package to calculate and plot growth curves of various readouts provided by Iris.

what can this tool do for me?

The script file kinetics.with.iris.R showcases a number of provided functions, including:

• reading iris files in a given folder, and parsing information in their filenames (see file load.all.iris.files.R)
• plotting the spatial bias of their values (see file plot.in.plate.format.3.R)
• calculating a colony's sigmoidal function fit for a given readout, and plotting the values and Gompertz fit to be saved as pdf (see file get.colony.growth.rates.2.R)

where can I get the kinetic tool?

you can download all scripts and example input and output files here

---

Acknowledgements

Iris makes use of the following open source projects:

• ImageJ
• FeatureJ
• gson
• opencsv

How does Iris work?

I thought you'd never ask. Lets take a look under the hood

How can I adjust Iris parameters?

This is documented in the next page: look under the hood

Support or Contact

Since Iris is still under active development, I would greatly appreciate any comments regarding it's performance and/or issues you might encounter.

You can get help on Iris-related questions or issues on the
Iris Q&A forum