Abstract A statistical approach using a polynomial linear model in combination with a probability distribution model was developed to mathematically represent the process of bacterial attachment and study its mechanism. The linear deterministic model was built based on data from experiments investigating bacterial and substratum surface physico-chemical factors as predictors of attachment. The prediction results were applied to a normal-approximated binomial distribution model to probabilistically predict attachment. The experimental protocol used mixtures of Streptococcus salivarius and Escherichia coli, and mixtures of porous poly(butyl methacrylate-co-ethyl dimethacrylate) and aluminum sec-butoxide coatings, at varying ratios, to allow bacterial attachment to substratum surfaces across a range of physico-chemical properties (including the surface hydrophobicity of bacterial cells and the substratum, the surface charge of the cells and the substratum, the substratum surface roughness and cell size). The model was tested using data from independent experiments. The model indicated that hydrophobic interaction was the most important predictor while reciprocal interactions existed between some of the factors. More importantly, the model established a range for each factor within which the resultant attachment is unpredictable. This model, however, considers bacterial cells as colloidal particles and accounts only for the essential physico-chemical attributes of the bacterial cells and substratum surfaces. It is therefore limited by a lack of consideration of biological and environmental factors. This makes the model applicable only to specific environments and potentially provides a direction to future modelling for different environments.

中文翻译：

---

一种模拟细菌附着于非生物表面的物理过程的统计方法

摘要 采用多项式线性模型与概率分布模型相结合的统计方法，以数学方式表示细菌附着过程并研究其机制。线性确定性模型是基于研究细菌和基质表面物理化学因素作为附着预测因子的实验数据建立的。将预测结果应用于正态近似二项式分布模型以概率预测依恋。实验方案使用唾液链球菌和大肠杆菌的混合物，以及多孔聚（甲基丙烯酸丁酯-co-二甲基丙烯酸乙酯）和仲丁醇铝涂层的混合物，以不同的比例，允许细菌通过一系列物理化学特性（包括细菌细胞和基质的表面疏水性、细胞和基质的表面电荷、基质表面粗糙度和细胞大小）附着到基质表面。该模型使用来自独立实验的数据进行测试。该模型表明疏水相互作用是最重要的预测因子，而一些因素之间存在相互的相互作用。更重要的是，该模型为每个因素建立了一个范围，在该范围内产生的依恋是不可预测的。然而，该模型将细菌细胞视为胶体颗粒，仅考虑细菌细胞和基质表面的基本物理化学属性。因此，它受到缺乏对生物和环境因素的考虑的限制。这使得模型仅适用于特定环境，并可能为未来不同环境的建模提供方向。