| Acknowledgement | 第7-9页 |
| Abstract | 第9-10页 |
| 摘要 | 第11-13页 |
| Abbreviations | 第13-17页 |
| Chapter 1 Introduction | 第17-32页 |
| 1 Protein subcellular localization prediction in plant | 第19-21页 |
| 1.1 The significance of protein subcellular localization prediction | 第19页 |
| 1.2 State of the art | 第19-20页 |
| 1.3 Problems | 第20-21页 |
| 2 Gap filling for reconstructions of metabolic network | 第21-24页 |
| 2.1 Dead-end definition | 第21-22页 |
| 2.2 Existing approaches | 第22-23页 |
| 2.3 Problems and our strategy | 第23-24页 |
| 3 Metabolic reconstruction in plant | 第24-27页 |
| 3.1 Current state of metabolic reconstruction | 第24-26页 |
| 3.2 The application of metabolic network in plants | 第26-27页 |
| 4 Opportunities and challenges for "big biological regulatory network" reconstruction of rice | 第27-29页 |
| 4.1 Public resources for rice | 第27-28页 |
| 4.2 Challenges | 第28-29页 |
| 5 Significan ces | 第29-32页 |
| 5.1 Conquering the bias in subcellular localization prediction for plant | 第29-30页 |
| 5.2 Improving the precision and efficiency of gap filling for metabolic model | 第30页 |
| 5.3 Merging GRNs, PPIs and GSMMs approaches into a single framework | 第30-31页 |
| 5.4 Rice biomolecule information base | 第31页 |
| 5.5 Summary | 第31-32页 |
| Chapter 2 PSI: A Comprehensive and Integrative approach for accurateplant subcellular localization prediction | 第32-54页 |
| 1 Motivation | 第32-33页 |
| 2 Data and Methods | 第33-40页 |
| 2.1 Experimental data and gold standard | 第33-34页 |
| 2.2 Assessment of subcellular localization predictors | 第34-38页 |
| 2.3 Singular value decomposition analysis | 第38页 |
| 2.4 Integration of predictors by group-voting | 第38-39页 |
| 2.5 Integration of predictors by artificial neural-network | 第39页 |
| 2.6 P-values as statistical inference for significance | 第39-40页 |
| 3 Results | 第40-50页 |
| 3.1 Prediction bias | 第40-41页 |
| 3.2 Community integration using group-voting | 第41-43页 |
| 3.3 Community integration using artificial neural-network | 第43-45页 |
| 3.4 Community integration outperforms individual predictors | 第45-46页 |
| 3.5 Wisdom of group-voting and neural network combination for subcellular localization prediction | 第46-47页 |
| 3.6 Webserver implementation | 第47-48页 |
| 3.7 Comparison between PSI and other individual predictors | 第48-49页 |
| 3.8 The applicability of PSI in other plants | 第49-50页 |
| 4 Discussions and Summary | 第50-54页 |
| 4.1 Why we need to employ the combination of group-voting and neural network? | 第50-52页 |
| 4.2 Influence of experimental data as input on result output | 第52页 |
| 4.3 Towards more powerful prediction | 第52-54页 |
| Chapter 3 DEF: An automated dead-end filling approach based onendosymbiosis simulation | 第54-75页 |
| I Motivation | 第54-55页 |
| 2 Data and Methods | 第55-59页 |
| 2.1 Endosymbiosis simulation | 第55-56页 |
| 2.2 Model optimization based on endosymbiosis simulation | 第56-58页 |
| 2.3 Assessment Procedure | 第58页 |
| 2.4 Gap filling for E. coli iJR904 | 第58-59页 |
| 3 Results | 第59-72页 |
| 3.1 Performance evaluation | 第59-65页 |
| 3.2 Gap filling for E.coli iJR904 based on KEGG database | 第65-69页 |
| 3.3 Comparison of DEF with other methods | 第69-70页 |
| 3.4 Webserver implementation | 第70-72页 |
| 4 Discussions | 第72-74页 |
| 4.1 The applicability of DEF | 第72-73页 |
| 4.2 Comparison of DEF with others | 第73-74页 |
| 5. Summary | 第74-75页 |
| Chapter 4 The organelle-focused proteomes and interactomes in rice | 第75-89页 |
| 1 Motivation | 第75-76页 |
| 2 Data and Methods | 第76-79页 |
| 2.1 Defining the organelle-focused proteomes of rice | 第76-77页 |
| 2.2 Construction of the organelle-focused interactomes of rice and their functions | 第77页 |
| 2.3 Enrichment analysis for the organdie-focused interactomes | 第77-78页 |
| 2.4 Motif analysis | 第78-79页 |
| 3 Results | 第79-88页 |
| 3.1 Capturing the organdie-focused proteome of rice | 第79页 |
| 3.2 The organelle-focused interactomes of rice | 第79-81页 |
| 3.3 Function overview of the organelle-focused interactomes using GO | 第81-84页 |
| 3.4 Motif analysis in the global interactome | 第84-85页 |
| 3.5 Motif analysis in each organdie interactome | 第85-88页 |
| 4 Summary | 第88-89页 |
| Chapter 5 Architecture of a fully compartmented multilayer regulatorynetwork in rice | 第89-116页 |
| 1 Motivation | 第89-90页 |
| 2 Data and Methods | 第90-97页 |
| 2.1 Data Resources | 第90-92页 |
| 2.2 Workflow | 第92-93页 |
| 2.3 Metabolic draft reconstruction | 第93-94页 |
| 2.4 Manual curation for metabolic draft | 第94-95页 |
| 2.5 Multilayer network construction | 第95-96页 |
| 2.6 Subcellular localization prediction | 第96-97页 |
| 3 Results | 第97-114页 |
| 3.1 Genome-scale Metabolic Reconstruction | 第97-102页 |
| 3.2 Organelle-focused omic-scale multilayer regulatory network construction of rice | 第102-105页 |
| 3.3 Chromosome regulatory patterns | 第105-107页 |
| 3.4 RiceNetDB: a rice biomolecular information base | 第107-111页 |
| 3.5 Maximizing insight by linking annotation, subcellular location and multilayer networks | 第111-114页 |
| 4 Summary | 第114-116页 |
| Chapter 6 Conclusions and Future Perspectives | 第116-119页 |
| 1 Conclusions | 第116-118页 |
| 2 Future Perspectives | 第118-119页 |
| References | 第119-129页 |
| Appendix 1 | 第129-134页 |
| Appendix 2 | 第134-144页 |
| Appendix 3 | 第144-145页 |
| Appendix 4 | 第145-160页 |
| Appendix 5 | 第160-173页 |
| Biography | 第173-175页 |
