AutoComet: A fully automated algorithm to quickly and accurately analyze comet assays

Area of Science:

• Biochemistry
• Genetics
• Computational Biology

Background:

• DNA damage is a hallmark of cancer and neurodegenerative diseases.
• Current methods for quantifying DNA damage, like comet assays, are time-consuming and require manual analysis.
• Existing open-source algorithms lack speed and accuracy for comet analysis.

Purpose of the Study:

• To develop a novel, open-source algorithm for automated comet assay analysis.
• To overcome the limitations of manual curation and slow processing times in DNA damage quantification.
• To provide a fast, accurate, and scalable solution for comet analysis.

Main Methods:

• Developed AutoComet, an automated algorithm for comet segmentation and tail parameter quantification.
• Implemented automated filtering based on comet size, intensity, and head-tail connectivity for improved accuracy.
• Validated AutoComet against manual curation and other open-source software.

Main Results:

• AutoComet significantly reduces analysis time by over tenfold (less than 3 seconds per comet).
• The algorithm demonstrates high segmentation accuracy by filtering out poorly formed comets.
• AutoComet successfully identified statistically significant differences in DNA damage levels between cell groups.

Conclusions:

• AutoComet offers a fast, unbiased, and accurate method for quantifying DNA damage via comet assays.
• The automation minimizes curator bias and enhances scalability for large datasets.
• This algorithm will significantly improve the detection and study of DNA damage in various diseases.