医学统计学（扬州大学）功能基因组合基因表达统计分析.pptVIP

Functional genomics and gene expression data analysis Bioinformatics tools for pre-genomicsequence data analysis Post-genomic vision Post-genomic vision Gene expression profiling.The rationale, what we would like and related problems A note of caution: Gene expression analysis using DNA microarrays Transforming images into data Normalisation The data Multiple array experiments. Unsupervised clustering methods:Useful for class discovery (we do not have any a priori knowledge on classes) An unsupervised problem: clustering of genes. Clustering of experiments:The rationale Clustering of experiments:The problems Supervised analysis.If we already have information on the classes, our question to the data should use it.Class prediction based on gene expression profiles: Gene selection.We are interested in selecting those genes showing differential expression among the classes studied. Gene selection and class discrimination Sorry... the data was a collection of random numbers labelled for two classes And, genes are not only related to discrete classes... Of predictors and molecular signatures Predictor of clinical outcome in breast cancer Information mining Information mining applications.1) use of biological information as a validation criteria Information mining of DNA array data. Allows quick assignation of function, biological role and subcellular location to groups of genes. Used to understand why genes differ in their expression between two different conditions Sources of information: Free text Curated terms (ontologies, etc.) Gene OntologyCONSORTIUM  The objective of GO is to provide controlled vocabularies for the description of the molecular function, biological process and cellular component of gene products. These terms are to be used as attributes of gene products by collaborating databases, facilitating uniform queries across them. The controlled vocabularies of terms are structured to allow both attribution and querying to be