Epilepsy is defined by the International League Against Epilepsy (ILAE) as a chronic disease characterized by a predisposition to the occurrence of epileptic seizures that affect around 50 million people according to WHO, with Colombia being the third largest Latin American country prevalence It is estimated that up to 40% of the different types of epilepsy are of genetic origin and gene expression may be directly or indirectly involved with neurobiological mechanisms that are responsible for therapeutic failure during the epileptogenesis process due to the interindividual variability of response to antiepileptic drugs, promoting the existence of phenotypically refractory patients and contributing to the emergence of drug resistance. This research work is a descriptive cross-sectional observational study consisting of a sample of 29 patients diagnosed with epilepsy that meets the inclusion criteria. A clinical and paraclinical characterization of the patients was carried out, and subsequently the specific genetic variants were identified to perform a bioinformatic approach through which possible target proteins and signaling pathways involved in the drug response mechanism were possible. The most frequent epilepsy was of a generalized type with 34.48% and the most frequent neurological comorbidity was the global developmental delay with 65.5%. In 20 patients, the result of the molecular study was abnormal, identifying 60 genetic variants of which 86.7%, that is, 52 variants were classified as VOUS and 36 explained the phenotype of the patients. In addition, 13.3%, that is, 8 variants, were classified as probably pathogenic variants and 7 of these explained the phenotype of the patients. Through the bioinformatic analysis of the candidate genes, create 15 biological networks, of which in 7 networks corresponding to the genes CACNA1H, CNTN2, TSC1, EPM2A, SCN1A, KCNQ3 and PRICKLE1 were found the possible devices for the response to medications and other potential therapeutic targets were proposed.