Monte Carlo Simulation Study Of Proton-induced Clustered DNA Damage

The radiation damage induced by ionizing radiation and its protection have always been one of the hot issues in the research of life science,especially radiation biology,radiology,and radiotherapy.The Deoxyribonucleic acid(DNA)molecule damage in the cell nucleus will affect the replication and transcription of genetic information.If the damage cannot be repaired in time,it will cause serious biological consequences such as gene mutation,chromosome aberration,and cell inactivation.On the one hand,scientists use molecular biology techniques to detect DNA fragments obtained by separating DNA from irradiated cells,and evaluate DNA strand breaks yields.On the other hand,with the rapid development of computer hardware and artificial intelligence such as machine learning,the use of Monte Carlo method to simulate the DNA molecular damage of radiation cells and complete the calculation of the DNA strand break yields has become a research hotspot in the field of biological dosimetry.Based on the Geant4-DNA Monte Carlo track structure platform,this thesis simulates the low-dose radiation of high-linear energy transfer(LET)protons,and reveals the pattern of clustered DNA damage of nuclear DNA,especially the indirect clustered DNA damage induced in the chemical process.The main work is as follows:1.The simulation of DNA molecular damage based on the atomic geometry model of the cell nucleus has shortcomings such as long calculation time and slow convergence speed.This thesis proposes a geometric model based on the damage probability density function: that is,the energy deposition points in the physical process and the hydroxyl radical(·OH)sites in the chemical process are sampled conditionally,which to determine the data set of the energy deposition points and ·OH sites falling in the sensitive areas of the DNA chain by sampling probability parameters.Then different damage probability functions are used to get the data set of the DNA damage points space distribution.Finally,the regulatory factor is determined according to the results of molecular biology tests,which to consider the influence of histones and compressed DNA molecules on the number of damage points.The simulation experiment results show that the geometric model dose not affect the spatial distribution of damage points,which is consistent with the results of other experiments and simulation methods.2.Because of the problem of large amount of indirect damage point data,the thesis uses density-based spatial clustering of applications with noise algorithm(DBSCAN)to determine the damage point density and calculate the yield of DNA strand breaks.And proposes to improve the KD-tree algorithm to be suitable for searching the damage point neighborhoods,which reduces the computational time complexity,and accelerate the simulation process of calculating DNA strand break yields.The simulation experiment results show that the average speed is twice as fast,and the average calculation time is reduced by ten days.3.DNA strand breaks induced by indirect effect dominate the early DNA damage due to radiation.In view of the current lack of research on indirect clustered DNA damage,this thesis proposes a method to calculate direct damage and indirect damage separately in the calculation of DNA strand breaks using DBSCAN.That is,the damage points generated by energy deposition and the damage points generated by ·OH are used to calculate the DNA strand break yields.Therefore,there is a complete quantitative evaluation about the yield and complexity of the indirect damaged DNA strand breaks.The classification of different single strand break(SSB)and double strand break(DSB)is also refined,specifically SSB+,DSB+ and DSB++,to reflect the complexity of chain breaks.The simulation experiment results show that the indirect effect of ·OH has a significant contribution to the increase in yield the complexity of SSB and DSB.With the increase of LET,the yield of indirect SSB gradually decreases,and the yield of DSB first decreases and then increases;The multiplicity of DSB clusters(damage points included in each DSB)and its yield increases with the increase of LET,which confirms the conclusion that DNA fragments are not randomly distributed.This work achieves more precise information on clustered DNA damage induced by proton radiation at the molecular level with high speed;it provides an essential and powerful research method for the study of radiation biological damage mechanism.