Prognosis of Dengue Hemorrhagic Fever Utilizing Human Genome Data and Machine Learning Technology

Dengue is of worldwide public health concern as it has become one of the most significant arthropod-borne diseases, affecting an estimated 4 billion individuals globally. Dengue infections range from asymptomatic, to the debilitating febrile illness dengue fever (DF), and to dengue hemorrhagic fever (DHF), which is severe and potentially life-threatening. The precise immunopathology of DENV infection has yet to be elucidated, however, it has been shown that host genetics influence infection and disease development. Single nucleotide polymorphisms (SNPs) offer early clinical identification and better understanding of disease progression. Our study utilizes a cohort of patients from Brazil, who are infected with Dengue, and who have developed either DF or DHF. Individuals were genotyped for 20 immune-relevant SNPs, including a core set of 13 SNPs recognized to be most influential in disease progression. A machine learning (ML) algorithm together with human genotyped data was utilized to identify the combination of SNPs that have the greatest utility in predicting risk for developing the severe dengue phenotype. From this study, dengue severity can be predicted based solely on human genomic markers