|Title||Systems biology and statistical data integration of ~omics data sets|
|Source||Wageningen University. Promotor(en): Richard Visser, co-promotor(en): Chris Maliepaard. - S.l. : s.n. - ISBN 9789461735843 - 177|
|Publication type||Dissertation, internally prepared|
|Keyword(s)||systeembiologie - statistische gegevens - gegevensanalyse - gegevens verzamelen - metabolomica - loci voor kwantitatief kenmerk - genomica - eiwitexpressieanalyse - solanum tuberosum - aardappelen - databanken - systems biology - statistical data - data analysis - data collection - metabolomics - quantitative trait loci - genomics - proteomics - solanum tuberosum - potatoes - databases|
|Categories||Molecular Databases / Data Processing, Database Management|
In this thesis quality traits of potato were related to different highly multivariate ~omics datasets containing information on proteins, primary and secondary metabolites and gene expression. The objectives were to explore and compare different statistical techniques that are able to quantify these relationships, and to identify components responsible for prediction of quality. We propose a strategy to integrate two or more of such datasets and to select subsets of predictive components. We used potato flesh colour as an example trait and identified metabolites and expressed genes that are associated with flesh colour. We identified two putative novel non-volatile glycosides of carotenoid-derived metabolites and a novel putative connection with the flavonoid pathway. From a gas chromatography data set we identified genetic factors underlying variation in primary metabolism and found the amino acid beta-alanine associated with starch content. Finally we performed an integrated analysis with gene expression, metabolites and proteomics data and present an approach to select a limited set of predictive genes, metabolites and proteins.