RESUMO
Identifying biomarker genes and characterizing interaction pathways with high-dimensional and low-sample size microarray data is a major challenge in computational biology. In this field, the construction of protein-protein interaction (PPI) networks using disease-related selected genes has garnered much attention. Support vector machines (SVMs) are commonly used to classify patients, and a number of useful tools such as lasso, elastic net, SCAD, or other regularization methods can be combined with SVM models to select genes that are related to a disease. In the current study, we propose a new Net-SVM model that is different from other SVM models as it is combined with L1/2-norm regularization, which has good performance with high-dimensional and low-sample size microarray data for cancer classification, gene selection, and PPI network construction. Both simulation studies and real data experiments demonstrated that our proposed method outperformed other regularization methods such as lasso, SCAD, and elastic net. In conclusion, our model may help to select fewer but more relevant genes, and can be used to construct simple and informative PPI networks that are highly relevant to cancer.
Assuntos
Modelos Biológicos , Neoplasias/metabolismo , Domínios e Motivos de Interação entre Proteínas , Máquina de Vetores de Suporte , Algoritmos , Biologia Computacional/métodos , Perfilação da Expressão Gênica/métodos , Redes Reguladoras de Genes , Marcadores Genéticos , Humanos , Neoplasias/genética , Análise de Sequência com Séries de Oligonucleotídeos/métodos , Mapas de Interação de ProteínasRESUMO
Gilbert's syndrome is suspected in patients with unconjugated hyperbilirubinemia caused by decreased activity of the UDP-glucuronosyltransferase 1A1 (UGT1A1) gene in the absence of abnormal liver function and hemolysis. The major genetic variants underlying Gilbert's syndrome are TATA-box repeats of the promoter region and exon 1 G211A of the coding region, particularly in Asians. The efficacy of DNA melting curve analysis, however, has not been established for the G211A mutation. For rapid and accurate molecular diagnosis of Gilbert's syndrome, DNA melting curve analysis was evaluated for its genotyping capability not only for TATA-box repeats of the UGT1A1 promoter, but also for G211A of UGT1A1 exon 1. TA repeats within the TATA-box sequence and the exon 1 G211A mutation of the UGT1A1 gene were analyzed by DNA melting curve analysis. To evaluate the assay reliability, direct sequencing or polyacrylamide gel electrophoresis was used as a comparative method. All homozygous and heterozygous polymorphisms of A(TA)7TAA within the TATA-box allele and of exon 1 G211A mutants of the UGT1A1 gene were successfully identified with DNA melting curve analysis. DNA melting curve analysis is, therefore, an effective molecular method for the rapid diagnosis of Gilbert's syndrome, as it detects not only TATA-box polymorphisms but also the exon 1 G211A mutation located within the UGT1A1 gene.