Simulation of Proton-Induced DNA Damage Patterns Using an Improved Clustering Algorithm.

Area of Science:

• Computational biology
• Radiation physics
• Molecular toxicology

Background:

• Simulating deoxyribonucleic acid (DNA) damage from radiation is computationally intensive.
• Existing traversal algorithms are slow and require significant computing resources.
• Understanding clustered DNA damage is crucial for predicting biological effects of radiation.

Purpose of the Study:

• To develop a faster and more accurate algorithm for simulating clustered DNA damage.
• To improve the calculation of DNA strand break probabilities.
• To analyze the indirect effects of hydroxyl radical attack on DNA following proton irradiation.

Main Methods:

• Implemented an improved "density-based spatial clustering of applications with noise" (DBSCAN) algorithm using a KD-tree approach.
• Calculated clustered DNA damage by considering spatial distributions of energy deposition and hydroxyl radical attack.
• Simulated indirect DNA damage induced by proton treatment at the molecular level.

Main Results:

• The improved DBSCAN algorithm achieves high accuracy and speed in calculating clustered DNA damage.
• Simulations were successfully run on a standard i7 quad-core CPU, reducing computational demands.
• The results for DNA strand breaks are consistent with existing experimental and simulation data.

Conclusions:

• The enhanced DBSCAN algorithm provides an efficient method for molecular-level DNA damage simulation.
• This approach can predict the impact of various proton energies and track structures on DNA.
• The findings contribute to a better understanding of radiation-induced DNA damage mechanisms.