BIOLOGICAL KNOWLEDGE DISCOVERY HANDBOOK: Preprocessing, Mining, and Postprocessing of Biological Data -- CONTENTS -- PREFACE -- CONTRIBUTORS -- SECTION I: BIOLOGICAL DATA PREPROCESSING -- PART A: BIOLOGICAL DATA MANAGEMENT -- 1 GENOME AND TRANSCRIPTOME SEQUENCE DATABASES FOR DISCOVERY, STORAGE, AND REPRESENTATION OF ALTERNATIVE SPLICING EVENTS -- 1.1 INTRODUCTION -- 1.2 SPLICING -- 1.2.1 Mechanism of Splicing -- 1.2.2 Regulation of Splicing -- 1.3 ALTERNATIVE SPLICING -- 1.3.1 Introduction to Alternative Splicing -- 1.3.2 Mechanism of Alternative Splicing -- 1.3.3 Regulation of Alternative Splicing -- 1.3.4 Evolution and Conservation of Splicing and Alternative Splicing -- 1.4 ALTERNATIVE SPLICING DATABASES -- 1.4.1 Genomic and Transcriptomic Sequence Analyses -- 1.4.2 Literature Overview of Various Alternative Splicing Databases -- 1.4.3 SDBs -- 1.5 DATA MINING FROM ALTERNATIVE SPLICING DATABASES -- 1.5.1 Implementation of dbASQ and Utility of SDBs -- 1.5.2 Identification of Transcript-Initial and Transcript-Terminal Variation -- ACKNOWLEDGMENTS -- WEB RESOURCES -- REFERENCES -- 2 CLEANING, INTEGRATING, AND WAREHOUSING GENOMIC DATA FROM BIOMEDICAL RESOURCES -- 2.1 INTRODUCTION -- 2.2 RELATED WORK -- 2.3 TYPOLOGY OF DATA QUALITY PROBLEMS IN BIOMEDICAL RESOURCES -- 2.4 CLEANING, INTEGRATING, AND WAREHOUSING BIOMEDICAL DATA -- 2.4.1 Lessons Learned from Integrating and Warehousing Biomedical Data on Liver Genes and Diseases -- 2.4.2 Data Quality-Aware Solutions -- 2.4.3 Biological Entity Resolution and Record Linkage -- 2.4.4 Ontology-Based Approaches -- 2.5 CONCLUSIONS AND PERSPECTIVES -- WEB RESOURCES -- REFERENCES -- 3 CLEANSING OF MASS SPECTROMETRY DATA FOR PROTEIN IDENTIFICATION AND QUANTIFICATION -- 3.1 INTRODUCTION -- 3.2 PREPROCESSING APPROACH FOR IMPROVING PROTEIN IDENTIFICATION -- 3.2.1 Existing Approaches.

3.2.2 New Dynamic Wavelet-Based Spectra Preprocessing Method -- 3.3 IDENTIFICATION FILTERING APPROACH FOR IMPROVING PROTEIN IDENTIFICATION -- 3.3.1 Existing Approaches -- 3.3.2 New Target-Decoy Approach for Improving Protein Identification -- 3.4 EVALUATION RESULTS -- 3.4.1 Evaluation of New Proprocessing Method -- 3.4.2 Evaluation of New Identification Filtering Method -- 3.5 CONCLUSION -- REFERENCES -- 4 FILTERING PROTEIN-PROTEIN INTERACTIONS BY INTEGRATION OF ONTOLOGY DATA -- 4.1 INTRODUCTION -- 4.2 EVALUATION OF SEMANTIC SIMILARITY -- 4.2.1 Gene Ontology -- 4.2.2 Survey of Semantic Similarity Measures -- 4.2.3 Correlation with Functional Categorizations -- 4.3 IDENTIFICATION OF FALSE PROTEIN-PROTEIN INTERACTION DATA -- 4.3.1 Classification Method -- 4.3.2 Accuracy of PPI Classification -- 4.3.3 Reliability of PPI Data -- 4.4 CONCLUSION -- REFERENCES -- PART B: BIOLOGICAL DATA MODELING -- 5 COMPLEXITY AND SYMMETRIES IN DNA SEQUENCES -- 5.1 INTRODUCTION -- 5.2 ARCHAEA -- 5.3 PATTERNS ON INDICATOR MATRIX -- 5.3.1 Indicator Matrix -- 5.3.2 Test Sequences -- 5.4 MEASURE OF COMPLEXITY AND INFORMATION -- 5.4.1 Complexity -- 5.4.2 Fractal Dimension -- 5.4.3 Entropy -- 5.5 COMPLEX ROOT REPRESENTATION OF DNA WORDS -- 5.5.1 Pseudorandom Sequence on Unit Circle -- 5.6 DNA WALKS -- 5.6.1 Walks on Pseudorandom and Deterministic Complex Sequences -- 5.6.2 Variance -- 5.7 WAVELET ANALYSIS -- 5.7.1 Haar Wavelet Basis -- 5.7.2 Discrete Haar Wavelet Transform -- 5.7.3 Haar Wavelet Coefficients and Statistical Parameters -- 5.7.4 Hurst Exponent -- 5.8 ALGORITHM OF SHORT HAAR DISCRETE WAVELET TRANSFORM -- 5.8.1 Clusters of Wavelet Coefficients -- 5.8.2 Cluster Analysis of Wavelet Coefficients of Complex DNA Representation -- 5.9 CONCLUSIONS -- REFERENCES -- 6 ONTOLOGY-DRIVEN FORMAL CONCEPTUAL DATA MODELING FOR BIOLOGICAL DATA ANALYSIS -- 6.1 INTRODUCTION.

6.2 DESCRIPTION LOGICS FOR CONCEPTUAL DATA MODELING -- 6.2.1 Generic Common Conceptual Data Model CMcom -- 6.2.2 EER, UML, and ORM in Terms of CMcom -- 6.3 EXTENSIONS -- 6.3.1 Ontology-Driven Modeling -- 6.3.2 More Expressive Languages -- 6.4 AUTOMATED REASONING AND BIOLOGICAL KNOWLEDGE DISCOVERY -- 6.4.1 Exploiting Automated Reasoning Services -- 6.4.2 Finding New Relationships and Classes by Using Instances -- 6.5 CONCLUSIONS AND OUTLOOK -- REFERENCES -- 7 BIOLOGICAL DATA INTEGRATION USING NETWORK MODELS -- 7.1 INTRODUCTION -- 7.1.1 Data Sources -- 7.1.2 Issues with Data Integration from Multiple Sources -- 7.2 BIOLOGICAL NETWORK MODELS -- 7.2.1 Sequence-Based Approach to Predict Interactions and Functional Links between Proteins -- 7.2.2 Graph-Theoretic and Probabilistic-Based Protein Interaction Network Models -- 7.2.3 Models in Genetic Interaction Networks -- 7.3 NETWORK MODELS IN UNDERSTANDING DISEASE -- 7.3.1 Interactome Network for Disease Prediction -- 7.3.2 Network Perturbation Due to Pathogens -- 7.3.3 Network View of Cancer -- 7.4 FUTURE CHALLENGES -- ACKNOWLEDGMENT -- REFERENCES -- 8 NETWORK MODELING OF STATISTICAL EPISTASIS -- 8.1 INTRODUCTION -- 8.2 EPISTASIS AND DETECTION -- 8.3 NETWORK -- 8.3.1 Fundamental Definitions -- 8.3.2 Notions on Connectivity -- 8.3.3 Vertex Centrality -- 8.3.4 Degree Distributions -- 8.3.5 Networks for Epistasis Studies -- 8.4 GENE-ASSOCIATION INTERACTION NETWORK -- 8.5 STATISTICAL EPISTASIS NETWORKS -- 8.5.1 Network Construction and Analysis -- 8.5.2 Observations -- 8.5.3 Implications -- 8.6 CONCLUDING REMARKS -- ACKNOWLEDGMENT -- REFERENCES -- 9 GRAPHICAL MODELS FOR PROTEIN FUNCTION AND STRUCTURE PREDICTION -- 9.1 INTRODUCTION -- 9.2 GRAPHICAL MODELS -- 9.2.1 Directed Graphical Model -- 9.2.2 Undirected Model (Markov Random Field) -- 9.2.3 Discriminative versus Generative Model -- 9.2.4 Sequential Model.

9.2.5 Maximum-Entropy Markov Models and Label Bias Problem -- 9.2.6 Conditional Random Field -- 9.2.7 Summary of Models and Available Resources -- 9.3 APPLICATIONS -- 9.3.1 Gene Prediction Using Conditional Random Fields -- 9.3.2 Protein Function Prediction Using Markov Random Fields -- 9.3.3 Application to Protein Tertiary Structure Prediction -- 9.4 SUMMARY -- ACKNOWLEDGMENTS -- REFERENCES -- PART C: BIOLOGICAL FEATURE EXTRACTION -- 10 ALGORITHMS AND DATA STRUCTURES FOR NEXT-GENERATION SEQUENCES -- 10.1 ALIGNERS -- 10.1.1 Hash-Based Aligners -- 10.1.2 Prefix-Based Aligners -- 10.1.3 Distributed Architectures -- 10.2 ASSEMBLERS -- 10.2.1 Greedy Assemblers: Seed and Extend -- 10.2.2 Overlap-Layout-Consensus Assemblers -- 10.2.3 DBG-Based Assemblers -- REFERENCES -- 11 ALGORITHMS FOR NEXT-GENERATION SEQUENCING DATA -- 11.1 INTRODUCTION -- 11.2 DEFINITIONS AND NOTATIONS -- 11.3 REAL: A READ ALIGNER FOR MAPPING SHORT READS TO A GENOME -- 11.3.1 Algorithm -- 11.3.2 Experimental Results -- 11.4 CREAL: MAPPING SHORT READS TO A GENOME WITH CIRCULAR STRUCTURE -- 11.4.1 Algorithm -- 11.4.2 Experimental Results -- 11.5 DYNMAP: MAPPING SHORT READS TO MULTIPLE CLOSELY RELATED GENOMES -- 11.5.1 Algorithm -- 11.5.2 Experimental Results -- 11.6 CONCLUSION -- REFERENCES -- 12 GENE REGULATORY NETWORK IDENTIFICATION WITH QUALITATIVE PROBABILISTIC NETWORKS -- 12.1 CENTRAL DOGMA: GENE EXPRESSION IN A CELL -- 12.2 MEASURING EXPRESSION LEVELS: MICROARRAY TECHNOLOGY -- 12.3 UNDERSTANDING GENE REGULATORY NETWORKS: BASIC CONCEPTS -- 12.3.1 Constructing GRNs from Microarray Data -- 12.3.2 Models for Reverse Engineering GRNs -- 12.4 BAYESIAN NETWORKS FOR LEARNING GRNs -- 12.4.1 Bayesian Networks -- 12.4.2 Dynamic Bayesian Networks -- 12.4.3 Learning GRNs -- 12.5 TOWARD QUALITATIVE MODELING OF GRNs -- 12.5.1 Motivating Factors for Using Qualitative Models -- 12.5.2 QPNs.

12.6 QPNs FOR GENE REGULATION -- 12.6.1 Dynamic QPNs -- 12.6.2 Approach -- 12.6.3 Computational Experiments and Results -- 12.7 SUMMARY AND CONCLUSIONS -- REFERENCES -- PART D: BIOLOGICAL FEATURE SELECTION -- 13 COMPARING, RANKING, AND FILTERING MOTIFS WITH CHARACTER CLASSES: APPLICATION TO BIOLOGICAL SEQUENCES ANALYSIS -- 13.1 INTRODUCTION -- 13.1.1 Ranking and Clustering Motifs -- 13.1.2 Ensemble Methods -- 13.1.3 Motif Representation -- 13.1.4 Problem Statement -- 13.2 MOTIFS WITH CHARACTER CLASSES: A CHARACTERIZATION -- 13.2.1 On Transitive Properties of Character Classes -- 13.2.2 Minimal Motifs and Motif Priority -- 13.3 FILTERING BY MEANS OF UNDERLYING MOTIFS -- 13.3.1 Algorithm for Filtering Set of Motifs into Its Underlying Representative Set -- 13.4 EXPERIMENTAL RESULTS AND DISCUSSION -- 13.5 CONCLUSION -- ACKNOWLEDGMENTS -- REFERENCES -- 14 STABILITY OF FEATURE SELECTION ALGORITHMS AND ENSEMBLE FEATURE SELECTION METHODS IN BIOINFORMATICS -- 14.1 INTRODUCTION -- 14.2 FEATURE SELECTION ALGORITHMS AND INSTABILITY -- 14.2.1 Categorization of Feature Selection Algorithm -- 14.2.2 Potential Causes of Feature Selection Instability -- 14.2.3 Remark on Feature Selection Instability -- 14.3 ENSEMBLE FEATURE SELECTION ALGORITHMS -- 14.3.1 Ensemble Based on Data Perturbation -- 14.3.2 Ensemble Based on Different Data Partitioning -- 14.3.3 Performance on Feature Selection Stability -- 14.3.4 Performance of Sample Classification -- 14.3.5 Ensemble Size -- 14.3.6 Some Key Aspects in Ensemble Feature Selection Algorithms -- 14.4 METRICS FOR STABILITY ASSESSMENT -- 14.4.1 Rank-Based Stability Metrics -- 14.4.2 Set-Based Stability Metrics -- 14.4.3 Threshold in Stability Metrics -- 14.4.4 Remark on Metrics for Stability Evaluation -- 14.5 CONCLUSIONS -- ACKNOWLEDGMENT -- REFERENCES.

15 STATISTICAL SIGNIFICANCE ASSESSMENT FOR BIOLOGICAL FEATURE SELECTION: METHODS AND ISSUES.