Nikjoo H, Uehara S, Wilson W E, Hoshi M, Goodhead D T
MRC, Radiation and Genome Stability Unit, Harwell, Oxfordshire, UK.
Int J Radiat Biol. 1998 Apr;73(4):355-64. doi: 10.1080/095530098142176.
A brief review is presented of the basic concepts in track structure and the relative merit of various theoretical approaches adopted in Monte-Carlo track-structure codes are examined. In the second part of the paper, a formal cluster analysis is introduced to calculate cluster-distance distributions.
Total experimental ionization cross-sections were least-square fitted and compared with the calculation by various theoretical methods. Monte-Carlo track-structure code Kurbuc was used to examine and compare the spectrum of the secondary electrons generated by using functions given by Born-Bethe, Jain-Khare, Gryzinsky, Kim-Rudd, Mott and Vriens' theories. The cluster analysis in track structure was carried out using the k-means method and Hartigan algorithm.
Data are presented on experimental and calculated total ionization cross-sections: inverse mean free path (IMFP) as a function of electron energy used in Monte-Carlo track-structure codes; the spectrum of secondary electrons generated by different functions for 500 eV primary electrons; cluster analysis for 4 MeV and 20 MeV alpha-particles in terms of the frequency of total cluster energy to the root-mean-square (rms) radius of the cluster and differential distance distributions for a pair of clusters; and finally relative frequency distribution for energy deposited in DNA, single-strand break and double-strand breaks for 10MeV/u protons, alpha-particles and carbon ions.
There are a number of Monte-Carlo track-structure codes that have been developed independently and the bench-marking presented in this paper allows a better choice of the theoretical method adopted in a track-structure code to be made. A systematic bench-marking of cross-sections and spectra of the secondary electrons shows differences between the codes at atomic level, but such differences are not significant in biophysical modelling at the macromolecular level. Clustered-damage evaluation shows: that a substantial proportion of dose ( 30%) is deposited by low-energy electrons; the majority of DNA damage lesions are of simple type; the complexity of damage increases with increased LET, while the total yield of strand breaks remains constant; and at high LET values nearly 70% of all double-strand breaks are of complex type.
简要回顾径迹结构的基本概念，并考察蒙特卡罗径迹结构编码中采用的各种理论方法的相对优点。在本文的第二部分，引入了一种形式聚类分析来计算聚类距离分布。
对总实验电离截面进行最小二乘拟合，并与各种理论方法的计算结果进行比较。使用蒙特卡罗径迹结构编码Kurbuc来检验和比较由玻恩-贝特、贾因-哈尔、格里津斯基、金-拉德、莫特和弗里恩斯理论给出的函数所产生的二次电子能谱。使用k均值法和哈蒂根算法对径迹结构进行聚类分析。
给出了以下数据：实验和计算的总电离截面；作为蒙特卡罗径迹结构编码中使用的电子能量函数的逆平均自由程（IMFP）；500 eV初级电子由不同函数产生的二次电子能谱；4 MeV和20 MeVα粒子根据总聚类能量与聚类均方根（rms）半径的频率以及一对聚类的微分距离分布进行的聚类分析；最后是10MeV/u质子、α粒子和碳离子在DNA中沉积能量、单链断裂和双链断裂的相对频率分布。
已经独立开发了许多蒙特卡罗径迹结构编码，本文给出的基准测试有助于更好地选择径迹结构编码中采用的理论方法。对二次电子截面和能谱的系统基准测试表明，这些编码在原子水平上存在差异，但在大分子水平的生物物理建模中，这种差异并不显著。聚类损伤评估表明：相当一部分剂量（30%）由低能电子沉积；大多数DNA损伤损伤类型简单；损伤的复杂性随着传能线密度（LET）的增加而增加，而链断裂的总产量保持不变；在高LET值时，几乎70%的双链断裂是复杂类型。
