# ML-Biomat
> Machine Learning and Multiscale Modeling for Biomaterials
> 机器学习与多尺度模拟 — 面向生物·材料·结构研究

A bilingual (EN/ZH) technical blog for graduate students in biology and materials science. Covers machine learning, multiscale simulation, Python tutorials, biomaterials mechanics, and wet-lab data processing.

## Site Structure
- Home: https://ml-biomat.com/
- English Articles: https://ml-biomat.com/en/
- 中文文章: https://ml-biomat.com/zh/
- Categories: https://ml-biomat.com/categories/
- FiberNet Library: https://ml-biomat.com/fibernet/
- About: https://ml-biomat.com/about/

## Categories
- Machine Learning / 机器学习: https://ml-biomat.com/categories/machine-learning/
- Multiscale Modeling / 多尺度模拟: https://ml-biomat.com/categories/multiscale-modeling/
- Bio-Materials-Structures / 生物·材料·结构: https://ml-biomat.com/categories/biomaterials/
- Python Tutorials / Python教程: https://ml-biomat.com/categories/python-tutorials/
- Wet-lab Data Processing / 湿实验数据处理: https://ml-biomat.com/categories/wet-lab-data/
- Research Notes / 研究笔记: https://ml-biomat.com/categories/research-notes/

## All Articles
- [EN] [Practical AFM Force Curve Analysis with Python: From Raw Data to Mechanical Properties](https://ml-biomat.com/posts/afm-force-curve-python/): A step-by-step guide to processing AFM force curves in Python: baseline correction, contact point detection, adhesion force, and Young's modulus fitting with Hertz/Sneddon models.
- [EN] [Hierarchical Superstructures in Biomaterials: Mechanical Design from Molecules to Macroscale](https://ml-biomat.com/posts/bio-hierarchical-structures/): A navigational guide to hierarchical superstructures in biomaterials: silk fibroin, collagen, bone, nacre, spider silk, and cellulose analyzed through their structure-mechanics relationships.
- [EN] [The Nature of Deep Learning](https://ml-biomat.com/posts/deep-learning-intro-ml/): A systematic introduction to deep learning: neurons, activation functions, backpropagation, regularization, and transfer learning, with practical guidance on when deep learning is the right tool in materials science.
- [EN] [Engineered Superstructures: Origami, Honeycomb, Auxetic and Lattice Design](https://ml-biomat.com/posts/engineered-superstructures/): A comprehensive review of engineered superstructures: origami, honeycomb, auxetic, helical, and lattice designs in 2D and 3D, covering geometry, mechanisms, parameters, and design rules.
- [EN] [Feature Engineering for Materials Science: From Experimental Logs to ML-Ready Data](https://ml-biomat.com/posts/feature-engineering-materials-ml/): How to transform scattered experimental data into structured feature matrices for ML, with Python examples.
- [EN] [Fiber Network Mechanics: Key Concepts for Biomaterials Research](https://ml-biomat.com/posts/fiber-network-mechanics-intro/): An introduction to the mechanical behavior of fiber-based biomaterials: network architecture, strain-stiffening, and structure-property relationships.
- [EN] [GROMACS Molecular Dynamics Tutorial: From Structure Preparation to Trajectory Analysis](https://ml-biomat.com/posts/gromacs-md-tutorial/): A complete hands-on tutorial for GROMACS molecular dynamics simulation, covering installation, system preparation, energy minimization, equilibration, production MD, and trajectory analysis (RMSD/RMSF/Rg/H-bonds) using lysozyme as a worked example.
- [EN] [Molecular Dynamics Basics: Your First LAMMPS Simulation](https://ml-biomat.com/posts/lammps-first-simulation/): A hands-on guide to running your first molecular dynamics simulation with LAMMPS: installation, input script structure, force fields, water box simulation, and trajectory analysis.
- [EN] [Common Classification Models in Machine Learning: From Logistic Regression to Gradient Boosting](https://ml-biomat.com/posts/ml-classification-models/): A systematic overview of common ML classification models, from logistic regression to gradient boosting, with principles, code, and comparison.
- [EN] [Machine Learning for Fiber Mechanics: From Data Preparation to Model Deployment](https://ml-biomat.com/posts/ml-predict-fiber-mechanics/): Step-by-step guide to predicting fiber biomaterial mechanical properties with ML: data cleaning, feature engineering, model selection, and uncertainty quantification.
- [EN] [A Practical Introduction to Multiscale Modeling for Biomaterials](https://ml-biomat.com/posts/multiscale-modeling-biomaterials-intro/): A practical overview of multiscale modeling approaches for fiber-based biomaterials: from molecular dynamics to continuum mechanics, with Python code examples.
- [EN] [NetworkX Fundamentals: Understanding, Building, and Analyzing Networks with Python](https://ml-biomat.com/posts/networkx-fundamentals/): A complete hands-on introduction to NetworkX: from graph theory fundamentals to network construction, analysis metrics, and visualization.
- [EN] [NumPy for Materials Research: From Zero to Productive](https://ml-biomat.com/posts/numpy-crash-course/): Learn NumPy from scratch with real materials science examples: loading experimental data, array operations, statistics, curve fitting, and batch processing.
- [EN] [Data Processing with Pandas: From Lab Measurements to Analysis-Ready Data](https://ml-biomat.com/posts/pandas-materials-data/): A practical guide to using Pandas for materials research data: loading multi-source files, cleaning, group statistics, and visualization.
- [EN] [A Complete Guide to Setting Up a Python Scientific Computing Environment: From Zero to Productivity](https://ml-biomat.com/posts/python-env-setup/): A comprehensive three-platform guide covering virtual environments, Conda and Mamba, JupyterLab, and VS Code integration for materials science and biology graduate students.
- [EN] [Python Tools for Scientific Computing in Materials Research](https://ml-biomat.com/posts/python-tools-materials-research/): A curated guide to Python libraries for materials science: from data loading with ASE/pymatgen to ML with scikit-learn, plus practical AFM/SEM data analysis scripts.
- [EN] [Common Regression Models in Machine Learning](https://ml-biomat.com/posts/regression-models-ml/): A systematic guide to linear regression, Ridge, Lasso, polynomial regression, SVR, decision trees, random forests, and gradient boosting for scientific and engineering applications, with complete code examples.
- [EN] [Scientific Data Visualization with Python: Publication-Ready Figures](https://ml-biomat.com/posts/scientific-visualization-python/): How to create publication-quality figures with Matplotlib and Seaborn: multi-panel layouts, color schemes, high-res export, and common chart types for materials research.
- [EN] [Practical Curve Fitting with SciPy: From Basics to Advanced Models](https://ml-biomat.com/posts/scipy-curve-fitting/): A comprehensive guide to curve fitting with SciPy for materials and biology researchers. Linear regression, exponential decay, Hertz indentation, viscoelastic models, with complete code and real examples.
- [EN] [Practical SEM Image Analysis for Biomaterials Research](https://ml-biomat.com/posts/sem-image-analysis/): How to analyze SEM images of fiber networks with Python: thresholding, fiber measurement, orientation analysis, and batch processing using scikit-image.
- [EN] [Silk Fibroin: From Molecular Structure to Mechanical Function](https://ml-biomat.com/posts/silk-fibroin-structure-mechanics/): How silk fibroin's hierarchical structure—from amino acid sequence to beta-sheet crystallites—dictates its remarkable mechanical properties and guides biomaterials design.
- [EN] [Understanding Stress-Strain Curves: A Materials Science Primer](https://ml-biomat.com/posts/stress-strain-curves/): A beginner-friendly guide to interpreting stress-strain curves: elastic modulus, yield strength, ultimate strength, toughness, and how these concepts apply to soft biomaterials.
- [ZH] [Python实战AFM力曲线分析：从原始数据到力学性能](https://ml-biomat.com/posts/afm-force-curve-python-zh/): 手把手教你用Python处理AFM力曲线数据：从基线校正、接触点检测、粘附力提取到Hertz与Sneddon模型拟合杨氏模量，涵盖完整代码实现、常见陷阱和批量处理策略，面向生物材料研究生的实战指南。
- [ZH] [生物材料的层级超结构：从分子到宏观的力学设计](https://ml-biomat.com/posts/bio-hierarchical-structures-zh/): 生物材料从分子到宏观尺度的层级超结构全面导览：深入丝素蛋白、胶原蛋白、骨、珍珠层、蜘蛛丝和纤维素等自然材料的跨尺度组装规律，解析其结构-力学关系及仿生设计启示。
- [ZH] [深度学习的本质：从特征工程到表示学习](https://ml-biomat.com/posts/deep-learning-intro-zh/): 从神经元、激活函数、反向传播到正则化与迁移学习，系统讲解深度学习的核心概念与训练技巧，帮助材料科学和生物学研究生判断何时该用深度学习、何时传统机器学习方法更合适。
- [ZH] [多尺度模拟在纤维生物材料中的应用入门](https://ml-biomat.com/posts/multiscale-modeling-biomaterials-intro-zh/): 从分子动力学到粗粒化再到连续介质力学，系统介绍多尺度模拟在纤维生物材料研究中的核心方法、主流工具和实战思路，含Python代码示例与完整计算工作流搭建指南。
- [ZH] [工程超结构力学综述：折纸、蜂巢、拉胀、扭转与点阵的二维三维设计](https://ml-biomat.com/posts/engineered-superstructures-zh/): 全面综述工程超结构的力学设计原理：涵盖折纸、蜂巢、拉涨、扭转和点阵等经典类型，深入解析从几何构型到力学机制、从关键参数到设计规则的完整逻辑链。
- [ZH] [材料科学中的特征工程：从实验数据到机器学习就绪数据](https://ml-biomat.com/posts/feature-engineering-materials-zh/): 如何将零散的实验数据转化为结构化特征矩阵：系统讲解分类编码、交互特征构造、缺失值处理和小样本特征选择，这是材料科学机器学习项目成败的关键第一步。
- [ZH] [纤维网络力学基础：生物材料研究的关键概念](https://ml-biomat.com/posts/fiber-network-mechanics-zh/): 纤维网络体系的力学行为系统入门：涵盖随机与非仿射网络架构、熵弹性与应变强化机制、结构-功能关系，以及实验表征与计算模拟方法，面向生物材料研究生的实用指南。
- [ZH] [GROMACS分子动力学模拟实战：从结构准备到轨迹分析](https://ml-biomat.com/posts/gromacs-md-tutorial-zh/): GROMACS分子动力学模拟完整实战教程：以溶菌酶为实例，手把手演示从结构准备、能量最小化、平衡态到生产运行的全流程，含RMSD、RMSF、Rg和氢键轨迹分析。
- [ZH] [分子动力学入门：你的第一个LAMMPS模拟](https://ml-biomat.com/posts/lammps-first-simulation-zh/): 手把手带你跑通第一个分子动力学模拟：从LAMMPS安装配置、输入脚本结构解析、力场选择到水盒子建模和轨迹分析，零基础可上手的完整入门指南。
- [ZH] [常用的机器学习分类模型：从逻辑回归到梯度提升](https://ml-biomat.com/posts/ml-classification-models-zh/): 系统梳理常用机器学习分类模型：从逻辑回归、KNN、SVM到决策树、随机森林和梯度提升，含核心原理对比、scikit-learn代码实现和应用场景分析。
- [ZH] [机器学习预测纤维材料力学性能：从数据准备到模型部署](https://ml-biomat.com/posts/ml-predict-fiber-mechanics-zh/): 手把手教你用机器学习预测纤维生物材料的力学性能：从数据清洗、特征工程、模型选择到超参数优化、交叉验证和结果解释，贯穿完整科研工作流的实战教程。
- [ZH] [NetworkX 基础：用 Python 理解、构建与分析网络](https://ml-biomat.com/posts/networkx-fundamentals-zh/): NetworkX完整入门教程：从图论基础概念到网络构建、中心性分析、社区检测和可视化，手把手教你用Python理解和分析复杂网络，面向生物与材料科学研究者。
- [ZH] [NumPy完全入门：实验室数据处理必备](https://ml-biomat.com/posts/numpy-crash-course-zh/): 面向材料科学和生物学研究的NumPy速成教程：数组创建与运算、实验数据加载、统计分析与曲线拟合，用真实科研数据快速上手Python科学计算。
- [ZH] [Pandas实验数据处理实战：从原始数据到统计分析](https://ml-biomat.com/posts/pandas-materials-data-zh/): 面向材料科学与生物学研究的Pandas数据处理实战教程：多源数据读取、清洗、分组统计、合并与可视化，从实验室原始数据到分析就绪表格的完整工作流。
- [ZH] [Python 科学计算环境完全搭建指南：从零到生产力](https://ml-biomat.com/posts/python-env-setup-zh/): 从虚拟环境原理到Conda/Mamba实战再到Jupyter与VS Code配置，三平台全覆盖的Python科学计算环境搭建指南，面向生物与材料科学研究生的从零到生产力完整教程。
- [ZH] [Python科研工具箱完全指南：材料科学与生物研究必备](https://ml-biomat.com/posts/python-tools-materials-research-zh/): 面向生物与材料研究者的Python科学工具全指南：从NumPy数据处理、Matplotlib出版级绘图到ASE原子模拟和skimage图像分析，覆盖科研全流程的实用工具链。
- [ZH] [常用的机器学习回归模型：从线性基准到梯度提升](https://ml-biomat.com/posts/regression-models-zh/): 系统讲解线性回归、岭回归、Lasso、多项式回归、SVR、决策树、随机森林和梯度提升在科学与工程中的应用，含完整Python代码示例和模型选择决策框架。
- [ZH] [Python科研绘图完全指南：从数据到发表级图表](https://ml-biomat.com/posts/scientific-visualization-python-zh/): 面向生物与材料研究者的Matplotlib科学绘图教程：从折线图、散点图到多面板布局和出版级配色方案，手把手教你将实验数据转化为发表就绪的高质量图表。
- [ZH] [SciPy曲线拟合完全指南：从线性回归到非线性建模](https://ml-biomat.com/posts/scipy-curve-fitting-zh/): 面向材料科学和生物学研究的SciPy曲线拟合全指南：从线性回归、指数衰减到Hertz压痕模型与粘弹性拟合，含完整代码实现、置信区间估计和模型选择策略。
- [ZH] [SEM图像分析实战：纤维形态的自动化测量](https://ml-biomat.com/posts/sem-image-analysis-zh/): 用Python自动化分析SEM纤维网络图像：系统讲解阈值分割、纤维直径提取、取向分布分析和批量处理策略，基于scikit-image的完整实战工作流与代码。
- [ZH] [丝素蛋白：从分子结构到力学功能](https://ml-biomat.com/posts/silk-fibroin-structure-mechanics-zh/): 丝素蛋白的层级结构与力学性能深度解析：从氨基酸序列到beta-折叠晶区的多尺度组装规律，以及其对生物材料理性设计的核心启示。
- [ZH] [应力-应变曲线解读：材料力学入门](https://ml-biomat.com/posts/stress-strain-curves-zh/): 面向生物与材料研究者的应力-应变曲线解读指南：从弹性模量、屈服强度、极限强度到韧性和粘弹性，系统理解材料力学行为的基本概念及其在软物质材料中的应用。