Abstract | ||
---|---|---|
Complex diseases, as Type 2 Diabetes, arise from dysfunctional complex biological mechanisms, caused by multiple variants on underlying groups of genes, combined with lifestyle and environmental factors. Thus far, the known risk factors are not sufficient to predict the manifestation of the disease. Genome-Wide Association Studies (GWAS) data were used to test for genotype-phenotype associations and were combined with a network-based analysis approach. Three datasets of genes associated with this disease were built and features were extracted for each of these genes. Machine learning models were employed to develop a predictor of the risk associated with Type 2 Diabetes to help the identification of new genetic markers associated with the disease. The obtained results highlight that the use of gene regions and protein-protein interaction networks can identify new genes and pathways of interest and improve the model performance, providing new possible interpretation for the biology of the disease. |
Year | DOI | Venue |
---|---|---|
2021 | 10.1007/978-3-030-86258-9_1 | PRACTICAL APPLICATIONS OF COMPUTATIONAL BIOLOGY & BIOINFORMATICS, PACBB 2021 |
DocType | Volume | ISSN |
Conference | 325 | 2367-3370 |
Citations | PageRank | References |
0 | 0.34 | 0 |
Authors | ||
5 |
Name | Order | Citations | PageRank |
---|---|---|---|
Débora Antunes | 1 | 0 | 0.34 |
Daniel Martins | 2 | 0 | 0.34 |
Fernanda Correia | 3 | 0 | 0.34 |
Miguel Rocha | 4 | 20 | 18.26 |
Joel P. Arrais | 5 | 0 | 1.35 |