Cheminformatics with Python

Machine learning and deep learning have now been widely used in cheminformatics, and programming skills are becoming a must for most chemists. Python has become an invaluable and highly popular open-source programming language that is ideally suited for data analysis and artificial intelligence in the field. Cheminformatics with Python provides a ground-up, practical introduction that will help the reader make effective use of the software, demonstrating how to use Python to write efficient cheminformatics programs and how to apply it to solve practical chemical problems. The book contains four main parts: programming, data, methods, and applications. In the programming section, a brief introduction to Python language and related scientific computing, cheminformatics, machine learning, and deep learning packages is provided, building knowledge from the ground up. In the data section, a systematic study of the representation of instrumental data, representation of molecular structures, and common chemical databases is given. In the methods section, analytical signal processing, multivariate calibration, multivariate resolution, classical machine learning, and deep learning methods are introduced in detail. The application section then looks at case studies of successful applications of cheminformatics in analytical chemistry, metabolomics, drug discovery, materials science, and other research areas which are demonstrated in detail. Finally, in the supporting appendix section, the necessary mathematical, statistical, and information theory-related theories in the main text are provided, and practical tips such as code editors and source code management are also included. Online coding materials on GitHub and an individual Jupyter notebook for each chapter further support practical learning. Cheminformatics with Python is written primarily for senior undergraduate students, graduate students, post-docs, and professors primarily in the field of computational and analytical chemistry who are harnessing AI, as well as those in medicinal and biochemistry or materials science applying cheminformatics in drug discovery, materials design, or metabolomics research.

• Familiarizes reader with well-established techniques in cheminformatics, providing an in-depth understanding of the application of deep learning in cheminformatics using Python software
• Simultaneously introduces the basic principles and implementations of deep learning algorithms, demonstrating how to apply deep learning models to chemical data for prediction and classification using Python
• Provides rich case studies and practical project examples to help readers apply what they have learned to real chemical problems and data
• Accompanied by an online GitHub repository with relevant Python code for each chapter
• Chapters also have an accompanying Jupyter Notebook containing relevant data, methods, and application examples, which can be run directly to get the results