Towards Artificial Intelligence in Chemistry via the Rational Design of Chemical Compounds and Automated Discovery

The course aims to present modern chemistry in a language accessible to specialists from other fields. In this course, we will discuss the application of machine learning to the search for new substances and materials with optimal properties based on existing scientific data (rational design), as well as examine studies that employ robotic systems for the automated exploration of new chemical reactions.

Course language: English

To participate in the course, students must have an English proficiency level of (upper) intermediate or higher.

The course is designed to reflect state‑of‑the‑art chemistry in a language accessible to specialists from other fields. The renaissance of machine learning and the advancement of data science have created the possibility for rational design — the targeted design of chemical substances with optimal properties using available data, without often‑expensive discovery by trial‑and‑error and serendipity. Intensive automation of chemical research drives us towards the discovery of new compounds outsourced to robots (“chemputer”), as was demonstrated with several proof‑of‑concept devices.

In this course, we will discuss applications of classic machine learning (ML) and artificial neural networks (ANN) in chemistry, the basics of cheminformatics, and chemical databases (including their applications and limitations). We will briefly review the latest notable research in medicinal chemistry, catalysis, materials science, and organic synthesis to highlight the applications of classic ML and ANN. Successful examples of automated discovery of chemical compounds by specialized robots will also be discussed, as well as applications of artificial intelligence in predicting chemical syntheses.

Jupyter Notebook, freely available Python libraries, and open‑source datasets will be used for interactive demonstrations. To fully understand the course, you should be able to write simple scripts in Python using Jupyter Notebook and be familiar with NumPy and scikit‑learn. Although the course is most relevant for chemists and students in fields related to chemistry (such as materials scientists, biologists, physicists, medical students, and geologists), other STEM students may also find it interesting.

Course Outline

The course program includes 18 sessions: 12 lectures and 6 seminars

Course language: English

Requirements for students:

• English proficiency at the (upper) intermediate level or higher;
• Ability to read specialized literature in English with the help of a dictionary;
• Basic knowledge of mathematical analysis and linear algebra (e.g., understanding of partial derivatives, integrals, systems of linear equations, vectors, and matrices);
• Knowledge of Python 3 syntax and basic data structures;
• Ability to use Jupyter Notebook;
• Familiarity with the basic capabilities of NumPy, SciPy, Matplotlib, and scikit‑learn.

Telegram

Занятия проводятся на Химическом факультете МГУ 

В программе курса 18 занятий: 12 лекций и 6 семинаров

Старт курса: с 16 сентября по пятницам с 16:50 до 18:30

Аудитория 337

Язык проведения курса: английский

Требования к студентам:

• Знание английского на уровне (upper) intermediate или выше;
• Умение читать специализированную литературу на английском со словарем;
• Знание математического анализа и линейной алгебры на базовом уровне (что такое частная производная, интеграл, система линейных уравнений, вектор, матрица);
• Знание синтаксиса языка Python 3 и основных структур данных;
• Умение использовать JupyterNotebook;
• Знание базовых возможностей NumPy, SciPy, Matplotlib и scikit-learn.

Форма записи на курс 

Telegram

Страница курса на Teach-In

Набор на курс 2022 года закрыт