[PSI]-CIC: A Deep-Learning Pipeline for the Annotation of Sectored Saccharomyces cerevisiae Colonies
2024

[PSI]-CIC: A Deep-Learning Pipeline for the Annotation of Sectored Saccharomyces cerevisiae Colonies

Sample size: 200 publication 10 minutes Evidence: moderate

Author Information

Author(s): Collignon Jordan, Naeimi Wesley, Serio Tricia R., Sindi Suzanne

Primary Institution: University of California, Merced

Hypothesis

The study aims to develop a computational pipeline to identify and characterize features of sectored yeast colonies.

Conclusion

[PSI]-CIC successfully segments and classifies yeast colonies, achieving high accuracy in identifying prion states and sector frequencies.

Supporting Evidence

  • The pipeline correctly predicts the state of approximately 95% of colonies detected.
  • The frequency of sectors in approximately 89.5% of colonies detected was accurately estimated.
  • The approach allows for more meaningful comparisons between experiments.

Takeaway

This study created a computer program that helps scientists quickly analyze yeast colonies to understand how they change color due to prions.

Methodology

The study used a U-Net architecture for image segmentation and classification of yeast colonies based on synthetic training data.

Potential Biases

Potential bias due to reliance on synthetic images for training, which may not fully capture the complexity of real experimental data.

Limitations

The model is highly tuned to specific image resolutions and may not generalize well to other datasets with different characteristics.

Digital Object Identifier (DOI)

10.1007/s11538-024-01379-w

Want to read the original?

Access the complete publication on the publisher's website

View Original Publication