I. Introduction
The problem of mental disorders is becoming exacerbated from year to year. Depression, which is the main subject of our current investigation, is considered one of the most significant mental disorders. In 2001 World Health Organization (WHO) noted the high dynamics of the growth of depression and suggested depression to become the second most severe public health problem in the world by the year 2020 [1], [2]. Since 2017 and until now WHO regards depression as the leading cause of disability and one of the key factors of suicide. For that reason, it is highly demanded to investigate mental disorders and the effective methods for their treatment. But researchers point out the insufficient progress in this direction [3], [4]. From our point of view, the more intensive implementation of informational technologies could solve this problem and improve the efficiency of scientific investigations, as was shown in other areas of biology and medicine. One of the dominant directions in the use of IT in medicine is machine learning, which has become widespread in oncology. Examples and principles of its implementation and the results are described in [5]. Explainable artificial intelligence methods, such as decision-making support systems based on ontologies, are also developed actively. Detailed information about such systems and the most significant medical and biological ontologies can be found in [6], [7]. However, in the area of mental disorders, the usage of such technologies is rather limited. We suppose that the primary reason for their unpopularity and inefficiency in this area is almost a total lack of commonly accepted standards and formal models, which could serve as a basis for the development and functioning of particular algorithms. Therefore, it is necessary to systematize notions and formalize this data domain. Let us consider now already existing attempts.