论文部分内容阅读
Backgroud: The rapid growth of high-throughput experimental data of biology is providing more and more valuable information on genome-wide molecular enrichment profiles in living cells or tissues.Inferring gene regulatory relationships from the highthroughput experimental data of different platforms can shed light on the mechanisms of biological networks implementing distinct functions.Methods: Here we proposed two different statistical models to assess two different regulatory relationships, respectively.We used a Lasso regression models to identify microRNA-target gene regulatory relationships.We developed another model combining Principal Component Analysis (PCA) and Supporting Vector Regression (SVR) to assess the regulatory functions ofhistone modification on gene expression.Results: By comparing the Lasso model with two other known methods applied to three different datasets, we found that the Lasso regression model has considerable advantages in both sensitivity and specificity.The regression coefficients in the model can be used to determine the true regulatory efficacies in tissues and was demonstrated using the miRNA target site type data.Finally, by constructing the miRNA regulatory networks in two stages of prostate cancer (PCa), we found the several significant miRNA-hubbed network modules associated with PCa metastasis.For the second model, we found the PCA+SVR model could significantly improve the predictive power of histone modification on gene expression while the numbers of predictors were the same.Conlusions: Both two statistical models demonstrated their powers of modeling the regulatory relationships using high-throughput experimental data.We anticipated that the models we constructed in this study can shed light on the studies of other types of regulatory relationships .