| dc.description.abstract | The Human Immunodeficiency Virus (HIV) is the causative agent of Acquired Immunodeficiency Syndrome (AIDS). The number of cases has been progressing and it has become a chronic disease thanks to the improvements in the therapeutic approaches we have available, a long period of clinical latency and persistent viral replication. But at the same time, there has been an increase in resistance and a growing need for new drugs with fewer side effects. This is why, in recent years, many researchers have been searching for new useful compounds against the disease. The use of machine learning for the development of new drugs is a booming field. This is how we have found the IFPTML methodology (IFPTML = IF + PT + ML) that brings together perturbation theory (PT), machine learning (ML) and information fusion (IF) methods. In this thesis, thanks to the application of IFPTML, an innovative approach for drug development is proposed, which includes the different phases of drug development (preclinical and clinical phases), generating a predictive model for anti-HIV drugs. The information used from preclinical, clinical and epidemiological trials has been extracted from three large databases. Firstly, to collect preclinical data, we have used the ChEMBL database, where preclinical data such as the structure of the compounds and the testing conditions of biological analyses are collected. Secondly, to collect clinical data from trials, used the ClinicalTrials.gov database has been used, which collects multiple variables from clinical trials (characteristics, intervention, patients, requirements, etc.). The application of this cheminformatics model, which allows predicting the anti-HIV effect of new drugs, according to the data collected in the three phases, allows the researcher or industry to reduce costs or time invested in drug development. | es_ES |
Back