MCount: An automated colony counting tool for high-throughput microbiology
Data files
Sep 23, 2024 version files 56.70 MB
-
96-well_plate.zip
12.81 MB
-
Codes_revised.zip
14.06 MB
-
README.md
1.48 KB
-
results.zip
29.82 MB
Abstract
Accurate colony counting is crucial for assessing microbial growth in high-throughput workflows. However, existing automated counting solutions struggle with the issue of merged colonies, a common occurrence in high-throughput plating. To overcome this limitation, we propose MCount, the only known solution that incorporates both contour information and regional algorithms for colony counting. By optimizing the pairing of contours with regional candidate circles, MCount can accurately infer the number of merged colonies. We evaluate MCount on a precisely labeled Escherichia coli dataset of 960 images (15,847 segments) and achieve an average error rate of 3.99%, significantly outperforming existing published solutions such as NICE (16.54%), AutoCellSeg (33.54%), and OpenCFU (50.31%). MCount is user friendly as it only requires two hyperparameters. To further facilitate deployment in scenarios with limited labeled data, we propose statistical methods for selecting the hyperparameters using few labeled or even unlabeled datapoints, all of which guarantee consistently low error rates. MCount presents a promising solution for accurate and efficient colony counting in application workflows requiring high throughput, particularly in cases with merged colonies.
https://doi.org/10.5061/dryad.2280gb62f
File: Codes.zip
Description: Includes .ipynb Python source code and example images / results.
File: 96-well plate.zip
Description: Contains 10 microplate images of the original resolution. Each filename is self-explanatory (an image was independently taken at a different condition or timepoint)
File: results.zip
Description: Contains a segmentation and quantified data from the microplate images necessary for replicating the study.
- Each subfolder “image_1”, “image 2”… “image 10” corresponds to data analyzed from the original image contained in folder “96-well plate.zip”
- In each subfolder (e.g., “results/image_1”), there are 96 JPG images with filenames ending with from “A1.jpg” to “H12.jpg”. They are partitions of the original image and refer to the well number of the microplate.
- In each subfolder, file named as “groundtruth.csv” refers to colony number that was manually counted by a human.
- In each subfolder, file named as “NICE.csv” refers to colony number that was analyzed by NICE software.
- In each subfolder, files named as “d_[xx] lambda [yy].csv” refer to colony number that was analyzed by MCount. In specific, [xx] refers to a value of hyperparameter “d” or “d_t”, and [yy] the value of λ (greek “lambda”) that were set for running the script.
Folder "96 well colonies" contains 10 microplate images with the original resolution. Folder "results" contains a segmentation and quantification results from the microplate images. Folder "Codes" includes Python source codes.