Source code transformation refers to modifications to the program's source code. These modifications are made in order to improve the quality of the code, to change the program behaviour, or to fix syntactic or semantic issues in programs. A number of reasons can necessitate to perform source code transformations. These may include learning about new requirements, changes to previous requirements, discovering malfunctions, poor readability of the code, and low efficiency of the program. In recent years, the use of deep learning-based algorithms to solve various software technology tasks has become increasingly popular. In many publications, we can read about methods that solve the task of source code transformation with algorithms based on deep learning.