09654nam 2200481 450 991063249100332120231110220838.03-031-20837-4(MiAaPQ)EBC7147179(Au-PeEL)EBL7147179(CKB)25483505500041(PPN)266348513(EXLCZ)992548350550004120230410d2022 uy 0engurcnu||||||||txtrdacontentcrdamediacrrdacarrierComputational intelligence methods for bioinformatics and biostatistics 17th international meeting, CIBB 2021, virtual event, November 15-17, 2021 : revised selected papers /edited by Davide Chicco [and seven others]Cham, Switzerland :Springer,[2022]©20221 online resource (269 pages)Lecture Notes in Computer Science ;v.13483Print version: Chicco, Davide Computational Intelligence Methods for Bioinformatics and Biostatistics Cham : Springer International Publishing AG,c2023 9783031208362 Includes bibliographical references and index.Intro -- Preface -- Organization -- Contents -- Chemical Neural Networks and Synthetic Cell Biotechnology: Preludes to Chemical AI -- 1 Can ``Synthetic Cell'' Biotechnology Become a Useful Platform for Chemical AI? -- 2 Scientific Background - What Exactly are SCs? -- 2.1 Computer Gestalt ch1varelabook vs. Autopoiesis &amp -- Autonomy -- 3 Bio-Chemical Neural Network -- 3.1 Selected Examples of Potentially Interesting CNNs for SCs -- 4 Concepts and Experimental Perspectives on Chemical Neural Networks and Synthetic Cells -- 4.1 Machine Learning -- 4.2 Meaning -- 4.3 Embodiment -- References -- Development of Bayesian Network for Multiple Sclerosis Risk Factor Interaction Analysis -- 1 Introduction -- 2 Previous Work -- 2.1 Artificial Intelligence (AI) and Machine Learning (ML) in MS Research -- 2.2 Alignment with Epidemiology -- 3 BN Development -- 3.1 Relevant Risk Factors -- 3.2 Structure -- 3.3 Measurements -- 4 Results and Discussion -- 4.1 Interaction, Sufficiency, Necessity -- 4.2 Equivalent Odds Ratios -- 5 Conclusions -- References -- Real-Time Automatic Plankton Detection, Tracking and Classification on Raw Hologram -- 1 Introduction -- 2 Materials and Methods -- 2.1 Hologram Formation -- 2.2 Dataset -- 2.3 Object Detection Models and Tracking -- 2.4 Metrics -- 3 Results -- 3.1 Detection Performances -- 3.2 Tracking Performances -- 4 Conclusion and Perspectives -- References -- The First in-silico Model of Leg Movement Activity During Sleep -- 1 Scientific Background -- 2 Materials and Methods -- 2.1 The LMA Model -- 2.2 Model Calibration -- 3 Results and Discussion -- 4 Conclusion -- References -- Transfer Learning and Magnetic Resonance Imaging Techniques for the Deep Neural Network-Based Diagnosis of Early Cognitive Decline and Dementia -- 1 Introduction -- 2 Deep Learning for Medical Diagnosis.2.1 Convolutional Neural Network for Image Classification -- 2.2 Pretrained Convolutional Neural Network -- 3 Imaging Data Repositories -- 4 Proposed Transfer Learning Pipeline -- 5 Experiments and Results -- 6 Discussion and Conclusions -- References -- Improving Bacterial sRNA Identification By Combining Genomic Context and Sequence-Derived Features -- 1 Background -- 2 Materials and Methods -- 2.1 Data -- 2.2 Feature Sets -- 2.3 Model Generation -- 2.4 Comparative Assessment -- 3 Results and Discussion -- 3.1 Model Selection -- 3.2 Variable Importance Analysis -- 3.3 Comparative Assessment -- 4 Conclusion -- References -- High-Dimensional Multi-trait GWAS By Reverse Prediction of Genotypes Using Machine Learning Methods -- 1 Background -- 2 Methods -- 2.1 Reverse Genotype and Trans-eQTL Prediction -- 2.2 Datasets -- 2.3 Experimental Settings -- 2.4 Code and Supplementary Information -- 3 Results -- 3.1 Reverse Genotype Prediction and Trans-EQTL Analysis in Simulated Data -- 3.2 Reverse Genotype Prediction and Trans-EQTL Analysis in Yeast -- 4 Discussion -- References -- A Non-Negative Matrix Tri-Factorization Based Method for Predicting Antitumor Drug Sensitivity -- 1 Background -- 2 Material and Methods -- 2.1 Datasets -- 2.2 Model -- 2.3 Method -- 2.4 Prediction of Novel Associations -- 2.5 Prediction of the Whole Drug Profile for a New Cell Line -- 3 Results -- 3.1 Prediction of Novel Associations -- 3.2 Prediction of the Whole Drug Profile for a New Cell Line -- 4 Discussion and Concluding Remarks -- References -- A Rule-Based Approach for Generating Synthetic Biological Pathways -- 1 Scientific Background -- 1.1 Introduction -- 1.2 Related Work -- 2 Materials and Methods -- 2.1 Synthetic Data Generation -- 2.2 Implementation Details -- 3 Experimental Setup -- 3.1 Model -- 3.2 Data -- 4 Results -- 4.1 Synthetic Data for Mixed-Batches.4.2 When to Use Synthetic Data -- 4.3 Generalizing to New Tasks -- 4.4 Computational Time -- 5 Conclusion -- References -- Machine Learning Classifiers Based on Dimensionality Reduction Techniques for the Early Diagnosis of Alzheimer's Disease Using Magnetic Resonance Imaging and Positron Emission Tomography Brain Data -- 1 Scientific Background -- 2 Methods -- 2.1 Dataset Description -- 2.2 Image Preprocessing -- 2.3 Feature Extraction -- 2.4 Dimensionality Reduction Techniques -- 2.5 Machine Learning Classifiers -- 2.6 Description of resampling Method and Performance Metrics -- 3 Result and Discussion -- 4 Conclusion -- References -- Text Mining Enhancements for Image Recognition of Gene Names and Gene Relations -- 1 Introduction -- 2 Related Work -- 3 Methods -- 3.1 Dataset -- 3.2 OCR Tool -- 3.3 Gene Name Enhancements -- 3.4 Gene Relation Enhancements -- 4 Results -- 4.1 Gene Name Enhancement Results -- 4.2 Gene Relation Enhancement Results -- 4.3 Use Cases -- 5 Discussion -- 6 Conclusion -- References -- Sentence Classification to Detect Tables for Helping Extraction of Regulatory Interactions in Bacteria -- 1 Introduction -- 2 Materials and Methods -- 2.1 Data Set -- 2.2 Feature Extraction and Vectorization -- 2.3 Supervised Learning -- 3 Results -- 3.1 Best Model -- 3.2 Best Features -- 4 Conclusion -- References -- RF-Isolation: A Novel Representation of Structural Connectivity Networks for Multiple Sclerosis Classification -- 1 Introduction -- 2 Materials and Methods -- 2.1 Study Population -- 2.2 MRI Acquisition and Processing -- 2.3 RF-Isolation Extraction -- 2.4 Classification Analysis -- 3 Results -- 3.1 Analysis of the MS-ProxIF Model -- 3.2 Comparison to Standard Network Measures -- 4 Conclusion -- References -- Summarizing Global SARS-CoV-2 Geographical Spread by Phylogenetic Multitype Branching Models -- 1 Introduction.2 Data and Methods -- 3 Results and Discussion -- 4 Conclusions -- References -- Explainable AI Models for COVID-19 Diagnosis Using CT-Scan Images and Clinical Data -- 1 Scientific Background -- 2 Materials and Methods -- 2.1 Datasets Description and Preprocessing -- 2.2 Models Design -- 2.3 Explainability and Interpretability -- 3 Results -- 3.1 Deep CNN for Image-Data Experimentation Results -- 3.2 Classifiers for Bio-Data Experimentation Results -- 3.3 Comparison Study -- 3.4 Explainability/Interpretability Results -- 4 Conclusion -- References -- The Need of Standardised Metadata to Encode Causal Relationships: Towards Safer Data-Driven Machine Learning Biological Solutions -- 1 Introduction -- 2 Considerations for the Development and Reporting of ML Solutions -- 2.1 The Desirable Properties of ML Models -- 2.2 Current Limitations in Biomedical ML Solutions -- 2.3 Origin and Error Types -- 2.4 Limitations of the Current Evaluation System -- 2.5 Helping Methodological Tools -- 3 Relevance of Induced Bias in Biological Studies for ML Analysis -- 4 An Approach to Overcome the Limitations: Accompanying Metadata with Causal Information -- 4.1 Incorporating Causal Information -- 5 Conclusion -- References -- Deep Recurrent Neural Networks for the Generation of Synthetic Coronavirus Spike Protein Sequences -- 1 Introduction -- 1.1 Coronaviridae -- 1.2 Recurrent Neural Networks -- 2 Methods -- 2.1 Recurrent Neural Network (RNN) Architecture -- 2.2 Coronavirus Training Set -- 3 Results -- 3.1 Characteristics of DL Simulated Spike Proteins -- 4 Conclusions -- References -- Recent Dimensionality Reduction Techniques for High-Dimensional COVID-19 Data -- 1 Introduction -- 2 State-of-the-Art Dimensionality Reduction Techniques -- 3 Experimental Analysis -- 3.1 Dataset Description and Preprocessing -- 3.2 Results -- 3.3 Discussion -- 4 Conclusions.References -- Soft Brain Ageing Indicators Based on Light-Weight LeNet-Like Neural Networks and Localized 2D Brain Age Biomarkers -- 1 Introduction -- 2 Methods -- 2.1 Data Extraction, Preprocessing and Labeling -- 2.2 2D-CNN Models for Brain Age Classification and Regression -- 2.3 2D Brain-Age Biomarkers Model Explanation -- 3 Results -- 3.1 Classification Results -- 3.2 Linear Regression Results -- 4 Discussion -- 5 Architectural, Qualitative and Performance Comparisons -- 6 Conclusion -- References -- Author Index.Lecture Notes in Computer Science Data miningData mining.943.005Chicco DavideMiAaPQMiAaPQMiAaPQBOOK9910632491003321Computational Intelligence Methods for Bioinformatics and Biostatistics774207UNINA