Animal & Avian Sciences Research Works

Permanent URI for this collectionhttp://hdl.handle.net/1903/1600

Browse

Author: search.filters.author.Zhang, Wensheng

Search Results

Now showing 1 - 2 of 2
  • Thumbnail Image
    Item
    Term-tissue specific models for prediction of gene ontology biological processes using transcriptional profiles of aging in drosophila melanogaster
    (Springer Nature, 2008-02-28) Zhang, Wensheng; Zou, Sige; Song, Jiuzhou
    Predictive classification on the base of gene expression profiles appeared recently as an attractive strategy for identifying the biological functions of genes. Gene Ontology (GO) provides a valuable source of knowledge for model training and validation. The increasing collection of microarray data represents a valuable source for generating functional hypotheses of uncharacterized genes. This study focused on using support vector machines (SVM) to predict GO biological processes from individual or multiple-tissue transcriptional profiles of aging in Drosophila melanogaster. Ten-fold cross validation was implemented to evaluate the prediction. One-tail Fisher's exact test was conducted on each cross validation and multiple testing was addressed using BH FDR procedure. The results showed that, of the 148 pursued GO biological processes, fifteen terms each had at least one model with FDR-adjusted p-value (Adj.p) <0.05 and six had the values between 0.05 and 0.25. Furthermore, all these models had the prediction sensitivity (SN) over 30% and specificity (SP) over 80%. We proposed the concept of term-tissue specific models indicating the fact that the major part of the optimized prediction models was trained from individual tissue data. Furthermore, we observed that the memberships of the genes involved in all the three pursued children biological processes on mitochondrial electron transport could be predicted from the transcriptional profiles of aging (Adj.p < 0.01). This finding may be important in biology because the genes of mitochondria play a critical role in the longevity of C. elegans and D. melanogaster.
  • Thumbnail Image
    Item
    Principal component tests: applied to temporal gene expression data
    (Springer Nature, 2009-01-30) Zhang, Wensheng; Fang, Hong-Bin; Song, Jiuzhou
    Clustering analysis is a common statistical tool for knowledge discovery. It is mainly conducted when a project still is in the exploratory phase without any priori hypotheses. However, the statistical significance testing between the clusters can be meaningful in helping the researchers to assess if the classification results from implementing a clustering algorithm need to be improved, even after the cluster number has been determined by a well-established criterion. This is important when we want to identify highly-specific patterns through classification. We proposed to use a principal component (PC) test, which is an implementation of an exact F statistic for the measures at multiple endpoints based on elliptical distribution theory, to assess the statistical significance between clusters. A challenge in the implementation is the choice of the number (q) of principal components to be considered, which can severely influence the statistical power of the method. We optimized the determination via validation according to a permutation test based on the clustering to be evaluated. The method was applied to a public dataset in classifying genes according to their temporal gene expression profiles. The results demonstrated that the PC testing were useful for determining the optimal number of clusters.