Abstract:

The aim of genome-wide association studies (GWAS) is the study of DNA variations, mainly the variations in the single nucleotyde polymorphisms. GWAS are employed to know how these variations affect certain traits or illness and, also, has proved its interest in order to know how patients behaves to certain pharmacological treatments which in future would make personalized medicine true.

Nowadays, GWAS are one of the most powerful tools available for researchers to understand human genetics. The first GWAS study was released in 2002 and a research published in 2009 is considered the first in this field to made use of machine learning methodologies. The use of this kind of methodologies has proved its interest in many areas of science and technology including health sciences.

This research presents the use of machine learning methodologies in the framework of doctoral thesis in the field of genome-wide association studies. On the one hand, the profile of students in this field makes challenging the use of this kind of methodologies as they require of a profound knowledge of the foundations of applied mathematics and statistics, on the other hand, the use of this techniques increases the interdisciplinary collaboration. In this research main machine learning methodologies applied to GWAS are detailed explaining how they must be employed and the challenges they suppose to doctoral students.

Keywords:

Genome wide association studies (GWAS), machine learning, doctoral thesis, health sciences.