Kooperativer Bibliotheksverbund

UID:

Format: 1 online resource (572 pages)

Edition: 1st ed.

ISBN: 9789811270611

Additional Edition: Print version: Altman, Russ B Biocomputing 2023 - Proceedings Of The Pacific Symposium Singapore : World Scientific Publishing Company,c2022 ISBN 9789811270604

Language: English

Keywords: Electronic books. ; Electronic books.

URL: Click to View

URL: Click here to view book

UID:

Format: 1 online resource (762 pages)

ISBN: 9789811215636

Note: Intro -- Content -- Preface -- ARTIFICIAL INTELLIGENCE FOR ENHANCING CLINICAL MEDICINE -- Session Introduction: Artificial Intelligence for Enhancing Clinical Medicine -- 1. Introduction -- 2. Novel Research Applying Artificial Intelligence to Clinical Medicine -- 2.1. Artificial intelligence for predicting patient outcomes -- 2.2. Artificial intelligence for improved insight into disease pathogenesis and features -- 3. Artificial intelligence for advancing medical workflows -- 4. Artificial intelligence for improving imaging -- 5. Conclusion -- References -- Predicting Longitudinal Outcomes of Alzheimer's Disease via a Tensor-Based Joint Classification and Regression Model -- 1. Introduction -- 2. Methods -- 2.1. The Longitudinal Joint Learning Model -- 2.2. The Solution Algorithm Using the Multi-Block ADMM -- 3. Experiments -- 3.1. Performance -- 3.2. Empirical Convergence -- 3.3. Biomarker Identification -- 4. Conclusion -- Acknowledgements -- References -- Robustly Extracting Medical Knowledge from EHRs: A Case Study of Learning a Health Knowledge Graph -- 1. Introduction -- 2. Related Work -- 3. Methods -- 3.1. Data collection and preparation -- 3.2. Evaluation with GHKG -- 3.3. Disease predictability analysis -- 3.4. Demographic analysis -- 3.5. Non-linear methods -- 4. Results -- 5. Discussion -- 5.1. Data size does not always matter. -- 5.2. Confounders may explain errors -- 5.3. Increased model complexity does not necessarily help -- 5.4. Limitations remain as an opportunity for future work -- 6. Conclusion -- Acknowledgements -- References -- Increasing Clinical Trial Accrual via Automated Matching of Biomarker Criteria -- 1. INTRODUCTION -- 2. MATERIALS AND METHODS -- 2.1. Specimens and Retrospective Analysis -- 2.2. Real-time Analysis -- 2.3. Source of Biomarker-based Clinical Trial Data -- 3. RESULTS. , 3.1. STAMP assay identifies somatic mutations -- 3.2. Algorithmic pipeline flags eligible patients for precision medicine clinical trials -- 3.2.1. Automation of Feature Matching -- 3.2.2. Manual Review of Matching Output -- 3.3. Validation of algorithmic pipeline -- 3.4. Match rate analysis of STAMP-identified mutations -- 4. DISCUSSION -- 4.1. Incorporation of informatics into clinical workflows -- 4.2. Limitations of algorithmic pipelines -- 5. CONCLUSION -- 6. AUTHOR CONTRIBUTIONS -- 7. ACKNOWLEDGEMENTS -- 8. REFERENCES -- 9. FIGURES -- 10. SUPPLEMENTARY TABLES AND FIGURES -- Addressing the Credit Assignment Problem in Treatment Outcome Prediction Using Temporal Difference Learning -- 1. Introduction -- 2. Dataset -- 3. Methods -- 3.1. Feature Extraction -- 3.2. Temporal Difference Learning -- 3.2.1. State-Estimation -- 3.2.2. Value Iteration -- 3.2.3. Optimization -- 3.3. Baselines and Performance Measure -- 4. Results -- 5. Discussion and Conclusion -- References -- Multiclass Disease Classification from Microbial Whole-Community Metagenomes -- 1. Introduction -- 2. Previous Work -- 3. Problem Setup -- 3.1. Dataset Construction -- 3.2. Graph Convolutional Neural Networks -- 3.3. Models -- 3.4. Training -- 4. Results -- 5. Conclusion -- 6. Acknowledgments -- 7. External Links -- References -- LitGen: Genetic Literature Recommendation Guided by Human Explanations -- 1. Introduction -- 2. Clinical Variant Curation Data -- 2.1. ClinGen's Variant Curation Interface (VCI) -- 2.2. Labeled papers -- 2.3. Unlabeled papers -- 3. Method -- 3.1. BiLSTM baseline -- 3.2. Leveraging unlabeled data -- 3.3. Explanations in multitask learning -- 3.4. Explanations as feature selection for proxy labeling -- 4. Experimental results -- 4.1. Evaluation metrics -- 4.2. Performance comparison -- 4.3. Performance of Proxy Labeling Model. , 4.4. Performance by Evidence Types -- 5. Discussion -- References -- From Genome to Phenome: Predicting Multiple Cancer Phenotypes Based on Somatic Genomic Alterations via the Genomic Impact Transformer -- 1. Introduction -- 2. Materials and methods -- 2.1. SGAs and DEGs pre-processing -- 2.2. The GIT neural network -- 2.2.1. GIT network structure: encoder-decoder architecture -- 2.2.2. Pre-training gene embeddings using Gene2Vec algorithm -- 2.2.3. Encoder: multi-head self-attention mechanism -- 2.2.4. Decoder: multi-layer perceptron (MLP) -- 2.3. Training and evaluation -- 3. Results -- 3.1. GIT statistically detects real biological signals -- 3.2. Gene embeddings compactly represent the functional impact of SGAs -- 3.4. Personalized tumor embeddings reveal distinct survival profiles -- 3.5. Tumor embeddings are predictive of drug responses of cancer cell lines -- 4. Conclusion and Future Work -- Acknowledgments -- Funding -- References -- Automated Phenotyping of Patients with Non-Alcoholic Fatty Liver Disease Reveals Clinically Relevant Disease Subtypes -- 1. Introduction -- 2. Methods -- 2.1. NAFLD definition -- 2.2. Natural language processing -- 2.3. Data collection -- 2.4. Clinical feature standardization and quality control -- 2.4.1. Demographic data -- 2.4.2. Diagnoses, procedures, medications -- 2.4.3. Laboratory tests -- 2.4.4. Vital signs -- 2.5. Patient pairwise distance and clustering -- 2.6. Statistical analysis -- 2.6.1. Descriptive statistics -- 2.6.2. Survival analysis -- 3. Results -- 3.1. Descriptive statistics for the cohort -- 3.2. Identification of NAFLD subtypes -- 3.3. Identification of distinct outcomes by NAFLD subtype -- 3.4. Internal cross-validation of the subtypes discovered -- 4. Conclusion -- 5. References -- References -- Monitoring ICU Mortality Risk with a Long Short-Term Memory Recurrent Neural Network. , 1. Introduction -- 2. Background and Related Work -- 3. Data and Preprocessing -- 3.1. Data Source and Cohort Selection -- 3.2. Data Extraction and Preprocessing -- 4. Methodology -- 4.1. Mortality Monitoring Task -- 4.2. Average Pooling and Attention Mechanism -- 4.3. Recurrent Neural Network (RNN) -- 5. Experimental Design -- 5.1. Sampling Strategy -- 5.2. Baseline Model -- 5.3. Experimental Settings -- 6. Results and Analysis -- 6.1. Dimensionality Analysis -- 6.2. Prediction with Different Feature Representations -- 6.3. Interpreting Mortality of Learned Representation -- 7. Discussion and Conclusions -- References -- Multilevel Self-Attention Model and Its Use on Medical Risk Prediction -- 1. Introduction -- 2. Related Work -- 2.1. Future disease prediction -- 3. Methods -- 3.1. Terminology and Notation -- 3.2. Model Architecture -- 3.3. Self-attention Encoder Unit -- 3.4. Loss Function -- 4. Experiments -- 4.1. Source of Data -- 4.2. Dataset preprocessing -- 4.3. Implementation details -- 5. Results -- 5.1. Future disease prediction -- 5.2. Future cost prediction -- 5.3. Case study for the self-attention mechanism -- 6. Conclusion -- 7. Bibliography -- Identifying Transitional High Cost Users from Unstructured Patient Profiles Written by Primary Care Physicians -- 1. Introduction -- 2. Data -- 2.1. EMRPC -- 2.2. Total Healthcare Costs -- 2.3. Encoding of Ordinal Variables -- 2.4. Word Embeddings -- 3. Methods -- 3.1. Bag of Words -- 3.2. EmbEncode -- 3.3. Historical Baseline -- 3.4. Varying the Training Set -- 3.5. Varying the Evaluation Set -- 4. Results -- 5. Discussion -- Acknowledgments -- References -- Obtaining Dual-Energy Computed Tomography (CT) Information from a Single-Energy CT Image for Quantitative Imaging Analysis of Living Subjects by Using Deep Learning -- 1. Introduction -- 2. Methods -- 3. Results. , 4. Discussion and Conclusion -- 5. Acknowledge -- References -- INTRINSICALLY DISORDERED PROTEINS (IDPS) AND THEIR FUNCTIONS -- Session Introduction: On the Importance of Computational Biology and Bioinformatics to the Origins and Rapid Progression of the Intrinsically Disordered Proteins Field -- 1. Introduction -- 2. Computational prediction of IDPs and IDRs and their functions -- 3. Popularization of research on IDPs and IDRs -- 4. A Collection of Recent Papers on IDPs and IDRs -- References -- Many-to-One Binding by Intrinsically Disordered Protein Regions -- 1. Introduction -- 2. Results -- 2.1. Many-to-one binding datasets -- 2.2. Many-to-one binding profiles: independent and overlapping -- 2.3 Comparing VOR (with backbone only) and RMSΔASA Values -- 2.4. Selected many-to-one case studies -- 3. Discussion -- 4. Materials and Methods -- 4.1. Dataset preparation -- 4.2. MoRF sequence similarity -- 4.3. MoRF interface similarity -- References -- Disordered Function Conjunction: On the In-Silico Function Annotation of Intrinsically Disordered Regions -- 1. Introduction -- 2. Materials and Methods -- 2.1. Data collection -- 2.2. Computational workflow -- 2.2.1. Feature-based representation of protein regions -- 2.2.2. Prediction of protein region functions -- 2.2.3. Assessment of the function prediction and clustering -- 3. Results and Discussion -- 3.1. Prediction of individual functions of IDRs -- 3.2. IDRs described in multidimensional space form function-related clusters -- 3.3. Case studies -- 4. Conclusions -- Acknowledgments -- References -- De novo Ensemble Modeling Suggests that AP2-Binding to Disordered Regions Can Increase Steric Volume of Epsin but Not Eps15 -- 1. Introduction -- 2. Methods -- 2.1. Generation of structural ensembles -- 2.2. Filtering Epsin conformers to mimic the effect of Plasma membrane. , 2.3. Docking AP2α to the IDRs by superposition.

Additional Edition: Print version: Altman, Russ B Biocomputing 2020 - Proceedings Of The Pacific Symposium Singapore : World Scientific Publishing Company,c2019 ISBN 9789811215629

Language: English

Keywords: Electronic books. ; Electronic books.

URL: Click to View

URL: Click here to view book

URL: https://www.worldscientific.com/worldscibooks/10.1142/11698#t=toc

UID:

Format: 1 online resource (372 pages)

ISBN: 981-12-3270-9

Content: The Pacific Symposium on Biocomputing (PSB) 2021 is an international, multidisciplinary conference for the presentation and discussion of current research in the theory and application of computational methods in problems of biological significance. Presentations are rigorously peer reviewed and are published in an archival proceedings volume. PSB 2021 will be held on a virtual platform at http://psb.stanford.edu/ on January 5–7, 2021. Tutorials and workshops will be offered prior to the start of the conference. PSB 2021 will bring together top researchers from the US, the Asian Pacific nations, and around the world to exchange research results and address open issues in all aspects of computational biology. It is a forum for the presentation of work in databases, algorithms, interfaces, visualization, modeling, and other computational methods, as applied to biological problems, with emphasis on applications in data-rich areas of molecular biology. The PSB has been designed to be responsive to the need for critical mass in sub-disciplines within biocomputing. For that reason, it is the only meeting whose sessions are defined dynamically each year in response to specific proposals. PSB sessions are organized by leaders of research in biocomputing's "hot topics." In this way, the meeting provides an early forum for serious examination of emerging methods and approaches in this rapidly changing field.

Note: Intro -- Contents -- Preface -- ACHIEVING TRUSTWORTHY BIOMEDICAL DATA -- Session Introduction: Achieving Trustworthy Biomedical Data Solutions -- 1. Introduction -- 2. Preserving Privacy and Explaining Decisions of Artificial Intelligence -- 3. Sharing Genomic and Health Records -- 4. Deploying Digital Health Solutions -- 5. Crowdsourcing Healthcare -- 6. Considering the Bioethics -- 7. Anticipating the Future -- References -- Selection of Trustworthy Crowd Workers for Telemedical Diagnosis of Pediatric Autism Spectrum Disorder -- 1. Introduction -- 2. Methods -- 2.1. Clinically representative videos -- 2.2. Crowdsourcing task for Microworkers -- 2.3. Classifier to evaluate performance -- 2.4. Metrics evaluated -- 2.5. Prediction of crowd worker performance from metrics -- 3. Results -- 3.1. Correlation between metrics and probability of the correct class -- 3.2. Regression prediction of the mean probability of the correct class -- 4. Discussion and Future Work -- 5. Conclusion -- 6. Acknowledgments -- References -- Differential Privacy Protection Against Membership Inference Attack on Machine Learning for Genomic Data -- 1. Introduction -- 2. Related Work -- 3. Methods -- 3.1. Membership inference attack (MIA) -- 3.2. Di erential privacy (DP) -- 4. Experimental Setup -- 4.1. Dataset -- 4.2. Implementation of target models -- 4.3. Implementation of DP -- 4.4. Implementation of MIA -- 4.5. Evaluation metrics -- 5. Results -- 5.1. Vulnerability of target model against MIA without DP protection -- 5.2. Impact of privacy budget on the target model accuracy -- 5.3. E ectiveness of DP against MIA -- 5.4. E ect of model sparsity -- 6. Conclusion -- References -- Making Compassionate Use More Useful: Using Real-World Data, Real-World Evidence and Digital Twins to Supplement or Supplant Randomized Controlled Trials -- 1. Introduction. , 1.1 Compassionate use -- 1.2 Compassionate use during the pandemic -- 1.3 What is an RCT? -- 1.3 EA data and NDAs -- 2. Real-World Information -- 2.1 Real-world data in trials -- 2.2 Real-world data and real-world evidence -- 2.2 Real-world limitations -- 3.0 Making RWD Work -- 3.1 Digital twins -- 4.0 Conclusions -- References -- ADVANCED METHODS FOR BIG DATA ANALYTICS IN WOMEN'S HEALTH -- Session Introduction: Advanced Methods for Big Data Analytics in Women's Health -- 1. Introduction -- 2. Session Summary -- 2.1. Full-length papers -- 3. Discussion -- References -- Intimate Partner Violence and Injury Prediction from Radiology Reports -- 1. Introduction -- 2. Related Work -- 2.1. Intimate partner violence -- 2.2. Clinical prediction -- 2.3. Natural language processing -- 3. Dataset -- 3.1. IPV patient selection -- 3.2. Control group selection -- 3.3. Injury labels -- 3.4. Data cleaning -- 3.5. Demographic data -- 4. Methodology -- 4.1. Experiment setup -- 4.2. Models -- 4.3. Evaluation -- 4.3.1. Prediction and predictive features -- 4.3.2. Error analysis -- 4.3.3. Report-program date gap -- 5. Results -- 5.1. IPV and injury prediction and predictive features -- 5.2. Error analysis -- 5.3. Report-program date gap -- 6. Discussion and conclusion -- References -- Not All C-sections Are the Same: Investigating Emergency vs. Elective C-section deliveries as an Adverse Pregnancy Outcome -- 1. Background and Significance -- 2. Methods -- 2.1. Dataset characteristics -- 2.2. Identification of delivery outcomes -- 2.2.1. Cesarean section deliveries -- 2.2.2. Preterm birth, stillbirth, and multiple birth deliveries -- 2.3. Integration of data from encounter records -- 2.4. Generalized regression models -- 3. Results -- 3.1. Utilization of cesarean section codes -- 3.2. Admission types recorded in encounter records. , 3.3. Age distribution by delivery admit type -- 3.4. Number of deliveries by weekday and admit type -- 4. Generalized regression model -- 4.1. Surgical Incision Type for C-section and Effect on Emergency Admission -- 5. Discussion -- References -- Co-occurrence Patterns of Intimate Partner Violence -- 1. Introduction -- 2. Materials and Methods -- 2.1. Description of Data and Pre-Processing -- 2.2. Co-Occurrence of Violence Types -- 2.3. Co-Occurrence Network of Individual Violence Items -- 2.4. Radial Visualization -- 2.5. Clustering of Survivors and Identification of Subgroups -- 2.6. Health Problems and Trauma Symptoms -- 3. Results -- 4. Discussion -- 5. Acknowledgments -- References -- BIOCOMPUTING AND AI FOR INFECTIOUS DISEASE MODELLING AND THERAPEUTICS -- Session Introduction: AI for Infectious Disease Modelling and Therapeutics -- 1. Background -- 2. Introduction -- 3. Social Media and COVID-19 -- 4. Biomedical literature and COVID-19 plus neglected tropical diseases -- 5. Genomics and HCV -- 6. Protein intrinsically disordered regions and SARS-CoV-2 -- 7. Protein-protein interactions and SARS-CoV-2 -- References -- Characterization of Anonymous Physician Perspectives on COVID-19 Using Social Media Data -- 1. Introduction -- 2. Methods -- 2.1. Data Collection -- 2.2. N-gram Frequency Measures -- 2.3. Sentiment Analysis -- 3. Results -- 3.1. Frequency of terms and n-grams -- 3.2. Sentiment analysis -- 3.3. Sentiments of tweets containing specific terms -- 4. Discussion and Conclusion -- 5. Acknowledgments -- References -- Semantic Changepoint Detection for Finding Potentially Novel Research Publications -- 1. Introduction -- 2. Methods -- 2.1. Data collection and general procedures -- 2.2. Title and abstract entropies -- 2.3. Bayesian changepoint analysis -- 2.4. Differential word clouds -- 2.5. Title and abstract embeddings. , 2.6. Semantic novelty -- 2.6.1. Strategy T1: Novel paper detection based on semantic distance -- 2.6.2. Strategy T2: Detection of novel papers that may constitute a trend -- 2.6.3. Strategy Y1: Detection of a group of novel papers based on their mean vector -- 2.6.4. Strategy Y2: Proportion of novel papers -- 3. Results and Discussion -- 4. Conclusions -- 5. Supplementary Information -- 6. Acknowledgements -- References -- TreeFix-TP: Phylogenetic Error-Correction for Infectious Disease Transmission Network Inference -- 1. Background -- 2. Methods -- 2.1. Minimizing inter-host transmissions -- 2.2. Description of TreeFix-TP -- 2.3. Evaluation using simulated data sets -- 2.3.1. Data set generation -- 2.3.2. Evaluating reconstruction accuracy -- 3. Results -- 3.1. Phylogenetic error correction results -- 3.2. Source recovery in HCV outbreaks -- 3.3. Running time and scalability -- 4. Discussion and Conclusions -- Acknowledgments -- Authors' Contributions -- Supplementary Material -- References -- SARS-CoV-2 Drug Discovery based on Intrinsically Disordered Regions -- 1. Introduction -- 2. Methods -- 2.1. Molecular docking -- 2.1.1. Data collection -- 2.1.2. Data preprocessing -- 2.1.3. Target file generation -- 2.1.4. Flexible docking -- 2.1.5. Ensemble docking -- 2.2. Statistical model -- 2.2.1. Chemprop -- 2.2.2. Data and training -- 3. Results -- 3.1. Interaction modelling -- 3.2. Activity prediction -- 4. Conclusion -- 5. Acknowledgements -- References -- Feasibility of the Vaccine Development for SARS-CoV-2 and Other Viruses Using the Shell Disorder Analysis -- 1. Introduction -- 1.1. SARS-COV-2 Vaccine -- 1.2. Shell disorder analysis of HIV and other viruses -- 1.3. Spinoff projects including coronaviruses: Shell disorder and modes of transmission -- 1.4. Yet another spinoff: Correlations between the inner shell disorder and virulence. , 2. Results -- 2.1. Clustering of CoV based mainly on NPID -- 2.2 Outer shell disorder is an indicator for the presence or absence of effective vaccines -- 2.3. A disordered outer shell provides an immune evasion tactic: Viral shapeshifting -- 2.4. SARS-CoV-2: Exceptionally hard shell (low MPID) associated with burrowing animals and buried feces -- 2.5. Behavior of the animal hosts matters in the evolutions of the viruses: EIAV vs. HIV -- 2.6. Feasibility of developing attenuated vaccine strains for SARS-CoV-2 -- 3. Discussion -- 3.1. Links between respiratory transmission, N (Inner shell) disorder, and virulence: Viral load in body fluids vs. vital organs -- 3.2. Greater disorder in the inner shell proteins provide means for the more efficient replication of viral particles -- 3.3 Two modes of immune evasion: "Trojan Horse" (inner shell disorder) and "viral shapeshifting" (outer shell disorder) -- 3.4. FIV, HIV-1 and HIV-2: Similarities and differences -- 3.5. FIV vaccine enigma: Questionable efficacy -- 4. Conclusions -- 4.1. Development of the SARS-CoV-2 vaccine is feasible and vaccine strains can be found in nature -- 5. Materials and Methods -- References -- Protein Sequence Models for Prediction and Comparative Analysis of the SARS-CoV-2−Human Interactome -- 1. Introduction -- 2. Methods -- 2.1. Generalized Additive Models with interactions (GA2M) -- 3. Gold Standard Interaction Datasets -- 3.1. Dealing with the lack of negative examples -- 3.2. Features -- 4. Experiments -- 4.1. TAPE: Transformer based model for protein sequences -- 5. Results -- 5.1. Prediction performance and validation of predicted interactions -- 5.2. Enrichment analysis of predicted human binding partners -- 6. Discussion -- 6.1. Visualizing the virus-human interactions -- 6.2. Highly ranked sequence features -- 6.3. Structural analysis -- 7. Prior Work -- 8. Conclusion. , 9. Acknowledgements. , English

Additional Edition: ISBN 981-12-3269-5

Language: English

UID:

Format: 1 Online-Ressource

ISBN: 9789811232701

Additional Edition: Erscheint auch als Druck-Ausgabe ISBN 978-981-123-269-5

Language: English

Subjects: Computer Science

Keywords: Biocomputer ; Konferenzschrift

DOI: 10.1142/12171

URL: Volltext  (kostenfrei)

URL: Volltext  (kostenfrei)

UID:

Format: 1 Online-Ressource

ISBN: 9789811215636

Additional Edition: Erscheint auch als Druck-Ausgabe ISBN 978-981-1215-62-9

Language: English

Subjects: Computer Science

Keywords: Biocomputer ; Konferenzschrift

URL: Volltext  (kostenfrei)

UID:

Format: 1 online resource (431 pages)

Note: Intro -- Content -- Preface -- AI-DRIVEN ADVANCES IN MODELING OF PROTEIN STRUCTURE -- Session Introduction: AI-Driven Advances in Modeling of Protein Structure -- 1. A short retrospect -- 2. A brief outline of current research -- 3. Future developments (complexes, ligand interactions, other molecules, dynamics, language models, geometry models, sequence design) -- 4. What is needed for further progress? -- 5. Overview of papers in this session -- 5.1. Evaluating significance of training data selection in machine learning -- 5.2. Geometric pattern transferability -- 5.3. Supervised versus unsupervised sequence to contact learning -- 5.4. Side chain packing using SE(3) transformers -- 5.5. Feature selection in electrostatic representations of ligand binding sites -- References -- Training Data Composition Affects Performance of Protein Structure Analysis Algorithms -- 1. Introduction -- 2. Methods -- 2.1. Experimental Design -- 2.2. Task-specific Methods -- 3. Results -- 3.1. Performance on NMR and cryo-EM structures is consistently lower than performance on X-ray structures, independent of training set -- 3.2. Inclusion of NMR data in the training set improves performance on held-out NMR data and does not degrade performance on X-ray data -- 3.3. Known biochemical and biophysical effects are replicated in trained models -- 3.4. Downsampling X-ray structures during training negatively affects performance on all types of data -- 4. Conclusion -- 5. Acknowledgments -- References -- Transferability of Geometric Patterns from Protein Self-Interactions to Protein-Ligand Interactions -- 1. Introduction -- 2. Related Work -- 3. Methods -- 3.1. Datasets -- 3.2. Contact extraction -- 3.3. Representing contact geometry -- 4. Results -- 4.1. Protein self-contacts exhibit clear geometric clustering. , 4.2. Many geometric patterns transfer to protein-ligand contacts -- 4.3. Application to protein-ligand docking -- 5. Conclusion and Future Work -- Supplemental Material, Code, and Data Availability -- Acknowledgments -- References -- Interpreting Potts and Transformer Protein Models Through the Lens of Simplified Attention -- 1. Introduction -- 2. Background -- 3. Methods -- 3.1. Potts Models -- 3.2. Factored Attention -- 3.3. Single-layer attention -- 3.4. Pretraining on Sequence Databases -- 3.5. Extracting Contacts -- 4. Results -- 5. Discussion -- Acknowledgements -- References -- Side-Chain Packing Using SE(3)-Transformer -- 1. Introduction -- 2. Methods -- 2.1. Neighborhood Graph Representation -- 2.2. The SE(3)-Transformer Architecture -- 2.3. Node Features -- 2.4. Final Layer -- 2.5. Rotamer Selection -- 2.6. Experiments -- 3. Results -- 4. Conclusion -- 5. Acknowledgements -- 6. References -- DeepVASP-E: A Flexible Analysis of Electrostatic Isopotentials for Finding and Explaining Mechanisms that Control Binding Specificity -- 1. Introduction -- 2. Methods -- 2.1. Convolutional Neural Network -- 2.2. Experimental Design -- 2.3. Comparison with Existing Methods -- 3. Results -- 4. Conclusions -- Acknowledgements -- References -- BIG DATA IMAGING GENOMICS -- Session Introduction: Big Data Imaging Genomics -- 1. Introduction -- 2. Overview of Contributions -- References -- A New Mendelian Randomization Method to Estimate Causal Effects of Multivariable Brain Imaging Exposures -- 1. Introduction -- 2. Methods -- 2.1. Step 1 : Mendelian randomization analysis on a single imaging exposure -- 2.2. Step 2: Joint instrumental variables and imaging exposures selection -- 2.3. Step 3: Causal effect identification for multiple imaging exposures -- 3. Application to evaluate the causal effect of white matter microstructure integrity on cognitive function. , 3.1. Data and study cohort -- 3.2. Results -- 4. Simulation -- 5. Discussion -- Funding -- Availability of data and materials -- Authors' contributions -- References -- Efficient Differentially Private Methods for a Transmission Disequilibrium Test in Genome Wide Association Studies -- 1. Introduction -- 2. Preliminaries -- 2.1. TDT -- 2.2. Differential Privacy -- 3. Methods -- 3.1. Exact Algorithm -- 3.2. Approximation Algorithm -- 4. Experiments -- 4.1. Simulation Data -- 4.2. Results -- 4.2.1. Run Time -- 4.2.2. Accuracy -- 4.3. Real Data -- 5. Conclusion -- Acknowledgement -- References -- Identifying Imaging Genetic Associations via Regional Morphometricity Estimation -- 1. Introduction -- 2. Methods -- 3. Materials -- 4. Experimental Design -- 5. Results and Discussion -- 6. Conclusion -- Acknowledgements -- References -- Identifying Highly Heritable Brain Amyloid Phenotypes Through Mining Alzheimer's Imaging and Sequencing Biobank Data -- 1. Introduction -- 2. Method -- 3. Materials -- 4. Experimental Workow -- 5. Results and Discussion -- 6. Conclusion -- Acknowledgements -- References -- Effects of ApoE4 and ApoE2 Genotypes on Subcortical Magnetic Susceptibility and Microstructure in 27,535 Participants from the UK Biobank -- 1. Introduction -- 2. Methods -- 2.1. UK Biobank Participants -- 2.2. T1-Weighted MRI -- 2.3. Quantitative Magnetic Susceptibility -- 2.4. Diffusion-Weighted MRI -- 2.5. Statistical Analyses -- 3. Results -- 3.1. ApoE4 Microstructural Associations -- 3.2. ApoE2 Microstructural Associations -- 3.3. ApoE-by-Age Interactions -- 3.3.1. ApoE Associations Stratified by Age -- 4. Discussion -- References -- Separating Clinical and Subclinical Depression by Big Data Informed Structural Vulnerability Index and Its impact on Cognition: ENIGMA Dot Product -- 1. Introduction -- 2. Methods. -- 2.1 Participants. , 2.2 Major Depressive Disorder Classification -- 2.3 Imaging Protocol and Processing -- 2.4 Calculation of linear indices of similarity -- 2.5 Calculation of QRI -- 2.7 Cognitive assessment -- 2.8 Statistics -- 3. Results -- 3.1 Group differences in symptoms and biomarkers -- 3.2 Effects of MDD on cognition. -- 3.3. Cognitive association -- 4. Discussion. -- 5. Conclusion -- 6. Acknowledgement -- References -- Generalizing Few-Shot Classification of Whole-Genome Doubling Across Cancer Types -- 1. Introduction -- 2. Related Work -- 3. Cohort -- 3.1. Cohort Selection -- 3.2. Feature Extraction -- 4. Methods -- 4.1. Model -- 4.2. Training -- 4.2.1. Pre-Training -- 4.2.2. Meta-Training -- 4.3. Meta-Validation and Meta-Test -- 4.4. Experiments -- 4.4.1. Cancer Types -- 4.4.2. Batch Effects -- 5. Results -- 5.1. Cancer Types -- 5.2. Batch Effects -- 5.2.1. Image Resolution -- 5.2.2. Image Brightness -- 6. Discussion -- Software and Data -- References -- HUMAN INTRIGUE: META-ANALYSIS APPROACHES FOR BIG QUESTIONS WITH BIG DATA WHILE SHAKING UP THE PEER REVIEW PROCESS -- Session Introduction: Human Intrigue: Meta-Analysis Approaches for Big Questions with Big Data While Shaking Up the Peer Review Process -- 1. Introduction -- 2. The Crowd Peer Review Process -- 2.1 Reviewer's Feedback -- 2.2 Conclusions -- 3. Meta-Analysis in Biocomputing -- 3.1 Novel Methods for Meta-Analysis of 'Omics Data -- 3.2 Using Publicly Available Data in Methods Development -- 3.3 Studying the Structure of Publicly Available Data -- 3.4 Conclusions -- Acknowledgements -- References -- Multitask Group Lasso for Genome Wide Association Studies in Diverse Populations -- 1. Introduction -- 2. Methods -- 2.1. Population stratification -- 2.2.1. Adjacency-constrained hierarchical clustering -- 2.2.2. LD-groups across populations -- 2.3. Multitask group Lasso. , 2.3.1. General framework and problem formulation -- 2.3.2. Related work -- 2.3.3. Gap safe screening rules -- 2.4. Stability selection -- 3. Experiments -- 3.1. Data -- 3.2. Preprocessing -- 3.3. Comparison partners -- 4. Results -- 4.1. MuGLasso draws on both LD-groups and the multitask approach to recover disease SNPs -- 4.2. MuGLasso provides the most stable selection -- 4.3. MuGLasso selects both task-speci c and global LD-groups -- 5. Discussion and Conclusions -- Acknowledgments -- Supplementary Materials and code -- References -- Mixed Effects Machine Learning Models for Colon Cancer Metastasis Prediction Using Spatially Localized Immuno-Oncology Markers -- 1. Introduction -- 2. Motivation for Comparison Study -- 2.1. Review of Prior Spatial Omics Analysis Methods -- 2.2. Motivation for Mixed Effects Machine Learning Approaches -- 3. Materials and Methods -- 3.1. Data Acquisition and Preprocessing -- 3.2. Experimental Design: Prediction Tasks and Modeling Approaches -- 4. Results -- 4.1. Macro: Inter-Tumoral Prediction -- 4.2. METS: Nodal and Distant Metastasis Prediction -- 5. Discussion -- 6. Conclusion -- 7. Acknowledgements -- 8. References -- Improving QSAR Modeling for Predictive Toxicology Using Publicly Aggregated Semantic Graph Data and Graph Neural Networks -- 1. Introduction -- 2. Methods -- 2.1. Obtaining toxicology assay data -- 2.2. Aggregating publicly available multimodal graph data -- 2.3. Heterogeneous graph neural network -- 2.3.1. Node classification -- 2.4. Baseline QSAR classifiers -- 3. Results -- 3.1. GNN node classification performance vs. baseline QSAR models -- 3.2. Ablation analysis of graph components' inuence on the trained model -- 4. Discussion -- 4.1. GNNs versus traditional ML for QSAR modeling -- 4.2. Interpretability of GNNs in QSAR -- 4.3. Sources of bias and their effects on QSAR for toxicity prediction. , 5. Conclusions.

Additional Edition: ISBN 981-12-5046-4

Additional Edition: ISBN 981-12-5047-2

Language: English

UID:

Format: 1 online resource (431 p.)

ISBN: 9789811250477 , 9811250472

Note: Description based upon print version of record. , 1. Introduction. , Intro -- Content -- Preface -- AI-DRIVEN ADVANCES IN MODELING OF PROTEIN STRUCTURE -- Session Introduction: AI-Driven Advances in Modeling of Protein Structure -- 1. A short retrospect -- 2. A brief outline of current research -- 3. Future developments (complexes, ligand interactions, other molecules, dynamics, language models, geometry models, sequence design) -- 4. What is needed for further progress? -- 5. Overview of papers in this session -- 5.1. Evaluating significance of training data selection in machine learning -- 5.2. Geometric pattern transferability , 5.3. Supervised versus unsupervised sequence to contact learning -- 5.4. Side chain packing using SE(3) transformers -- 5.5. Feature selection in electrostatic representations of ligand binding sites -- References -- Training Data Composition Affects Performance of Protein Structure Analysis Algorithms -- 1. Introduction -- 2. Methods -- 2.1. Experimental Design -- 2.2. Task-specific Methods -- 3. Results -- 3.1. Performance on NMR and cryo-EM structures is consistently lower than performance on X-ray structures, independent of training set , 3.2. Inclusion of NMR data in the training set improves performance on held-out NMR data and does not degrade performance on X-ray data -- 3.3. Known biochemical and biophysical effects are replicated in trained models -- 3.4. Downsampling X-ray structures during training negatively affects performance on all types of data -- 4. Conclusion -- 5. Acknowledgments -- References -- Transferability of Geometric Patterns from Protein Self-Interactions to Protein-Ligand Interactions -- 1. Introduction -- 2. Related Work -- 3. Methods -- 3.1. Datasets -- 3.2. Contact extraction , 3.3. Representing contact geometry -- 4. Results -- 4.1. Protein self-contacts exhibit clear geometric clustering -- 4.2. Many geometric patterns transfer to protein-ligand contacts -- 4.3. Application to protein-ligand docking -- 5. Conclusion and Future Work -- Supplemental Material, Code, and Data Availability -- Acknowledgments -- References -- Interpreting Potts and Transformer Protein Models Through the Lens of Simplified Attention -- 1. Introduction -- 2. Background -- 3. Methods -- 3.1. Potts Models -- 3.2. Factored Attention -- 3.3. Single-layer attention , 3.4. Pretraining on Sequence Databases -- 3.5. Extracting Contacts -- 4. Results -- 5. Discussion -- Acknowledgements -- References -- Side-Chain Packing Using SE(3)-Transformer -- 1. Introduction -- 2. Methods -- 2.1. Neighborhood Graph Representation -- 2.2. The SE(3)-Transformer Architecture -- 2.3. Node Features -- 2.4. Final Layer -- 2.5. Rotamer Selection -- 2.6. Experiments -- 3. Results -- 4. Conclusion -- 5. Acknowledgements -- 6. References -- DeepVASP-E: A Flexible Analysis of Electrostatic Isopotentials for Finding and Explaining Mechanisms that Control Binding Specificity

Additional Edition: Print version: Altman, Russ B Biocomputing 2022 - Proceedings Of The Pacific Symposium Singapore : World Scientific Publishing Company,c2021 9789811250460

Language: English

Keywords: Electronic books.

URL: Click here to view book

UID:

Format: 1 online resource (431 pages)

Note: Intro -- Content -- Preface -- AI-DRIVEN ADVANCES IN MODELING OF PROTEIN STRUCTURE -- Session Introduction: AI-Driven Advances in Modeling of Protein Structure -- 1. A short retrospect -- 2. A brief outline of current research -- 3. Future developments (complexes, ligand interactions, other molecules, dynamics, language models, geometry models, sequence design) -- 4. What is needed for further progress? -- 5. Overview of papers in this session -- 5.1. Evaluating significance of training data selection in machine learning -- 5.2. Geometric pattern transferability -- 5.3. Supervised versus unsupervised sequence to contact learning -- 5.4. Side chain packing using SE(3) transformers -- 5.5. Feature selection in electrostatic representations of ligand binding sites -- References -- Training Data Composition Affects Performance of Protein Structure Analysis Algorithms -- 1. Introduction -- 2. Methods -- 2.1. Experimental Design -- 2.2. Task-specific Methods -- 3. Results -- 3.1. Performance on NMR and cryo-EM structures is consistently lower than performance on X-ray structures, independent of training set -- 3.2. Inclusion of NMR data in the training set improves performance on held-out NMR data and does not degrade performance on X-ray data -- 3.3. Known biochemical and biophysical effects are replicated in trained models -- 3.4. Downsampling X-ray structures during training negatively affects performance on all types of data -- 4. Conclusion -- 5. Acknowledgments -- References -- Transferability of Geometric Patterns from Protein Self-Interactions to Protein-Ligand Interactions -- 1. Introduction -- 2. Related Work -- 3. Methods -- 3.1. Datasets -- 3.2. Contact extraction -- 3.3. Representing contact geometry -- 4. Results -- 4.1. Protein self-contacts exhibit clear geometric clustering. , 4.2. Many geometric patterns transfer to protein-ligand contacts -- 4.3. Application to protein-ligand docking -- 5. Conclusion and Future Work -- Supplemental Material, Code, and Data Availability -- Acknowledgments -- References -- Interpreting Potts and Transformer Protein Models Through the Lens of Simplified Attention -- 1. Introduction -- 2. Background -- 3. Methods -- 3.1. Potts Models -- 3.2. Factored Attention -- 3.3. Single-layer attention -- 3.4. Pretraining on Sequence Databases -- 3.5. Extracting Contacts -- 4. Results -- 5. Discussion -- Acknowledgements -- References -- Side-Chain Packing Using SE(3)-Transformer -- 1. Introduction -- 2. Methods -- 2.1. Neighborhood Graph Representation -- 2.2. The SE(3)-Transformer Architecture -- 2.3. Node Features -- 2.4. Final Layer -- 2.5. Rotamer Selection -- 2.6. Experiments -- 3. Results -- 4. Conclusion -- 5. Acknowledgements -- 6. References -- DeepVASP-E: A Flexible Analysis of Electrostatic Isopotentials for Finding and Explaining Mechanisms that Control Binding Specificity -- 1. Introduction -- 2. Methods -- 2.1. Convolutional Neural Network -- 2.2. Experimental Design -- 2.3. Comparison with Existing Methods -- 3. Results -- 4. Conclusions -- Acknowledgements -- References -- BIG DATA IMAGING GENOMICS -- Session Introduction: Big Data Imaging Genomics -- 1. Introduction -- 2. Overview of Contributions -- References -- A New Mendelian Randomization Method to Estimate Causal Effects of Multivariable Brain Imaging Exposures -- 1. Introduction -- 2. Methods -- 2.1. Step 1 : Mendelian randomization analysis on a single imaging exposure -- 2.2. Step 2: Joint instrumental variables and imaging exposures selection -- 2.3. Step 3: Causal effect identification for multiple imaging exposures -- 3. Application to evaluate the causal effect of white matter microstructure integrity on cognitive function. , 3.1. Data and study cohort -- 3.2. Results -- 4. Simulation -- 5. Discussion -- Funding -- Availability of data and materials -- Authors' contributions -- References -- Efficient Differentially Private Methods for a Transmission Disequilibrium Test in Genome Wide Association Studies -- 1. Introduction -- 2. Preliminaries -- 2.1. TDT -- 2.2. Differential Privacy -- 3. Methods -- 3.1. Exact Algorithm -- 3.2. Approximation Algorithm -- 4. Experiments -- 4.1. Simulation Data -- 4.2. Results -- 4.2.1. Run Time -- 4.2.2. Accuracy -- 4.3. Real Data -- 5. Conclusion -- Acknowledgement -- References -- Identifying Imaging Genetic Associations via Regional Morphometricity Estimation -- 1. Introduction -- 2. Methods -- 3. Materials -- 4. Experimental Design -- 5. Results and Discussion -- 6. Conclusion -- Acknowledgements -- References -- Identifying Highly Heritable Brain Amyloid Phenotypes Through Mining Alzheimer's Imaging and Sequencing Biobank Data -- 1. Introduction -- 2. Method -- 3. Materials -- 4. Experimental Workow -- 5. Results and Discussion -- 6. Conclusion -- Acknowledgements -- References -- Effects of ApoE4 and ApoE2 Genotypes on Subcortical Magnetic Susceptibility and Microstructure in 27,535 Participants from the UK Biobank -- 1. Introduction -- 2. Methods -- 2.1. UK Biobank Participants -- 2.2. T1-Weighted MRI -- 2.3. Quantitative Magnetic Susceptibility -- 2.4. Diffusion-Weighted MRI -- 2.5. Statistical Analyses -- 3. Results -- 3.1. ApoE4 Microstructural Associations -- 3.2. ApoE2 Microstructural Associations -- 3.3. ApoE-by-Age Interactions -- 3.3.1. ApoE Associations Stratified by Age -- 4. Discussion -- References -- Separating Clinical and Subclinical Depression by Big Data Informed Structural Vulnerability Index and Its impact on Cognition: ENIGMA Dot Product -- 1. Introduction -- 2. Methods. -- 2.1 Participants. , 2.2 Major Depressive Disorder Classification -- 2.3 Imaging Protocol and Processing -- 2.4 Calculation of linear indices of similarity -- 2.5 Calculation of QRI -- 2.7 Cognitive assessment -- 2.8 Statistics -- 3. Results -- 3.1 Group differences in symptoms and biomarkers -- 3.2 Effects of MDD on cognition. -- 3.3. Cognitive association -- 4. Discussion. -- 5. Conclusion -- 6. Acknowledgement -- References -- Generalizing Few-Shot Classification of Whole-Genome Doubling Across Cancer Types -- 1. Introduction -- 2. Related Work -- 3. Cohort -- 3.1. Cohort Selection -- 3.2. Feature Extraction -- 4. Methods -- 4.1. Model -- 4.2. Training -- 4.2.1. Pre-Training -- 4.2.2. Meta-Training -- 4.3. Meta-Validation and Meta-Test -- 4.4. Experiments -- 4.4.1. Cancer Types -- 4.4.2. Batch Effects -- 5. Results -- 5.1. Cancer Types -- 5.2. Batch Effects -- 5.2.1. Image Resolution -- 5.2.2. Image Brightness -- 6. Discussion -- Software and Data -- References -- HUMAN INTRIGUE: META-ANALYSIS APPROACHES FOR BIG QUESTIONS WITH BIG DATA WHILE SHAKING UP THE PEER REVIEW PROCESS -- Session Introduction: Human Intrigue: Meta-Analysis Approaches for Big Questions with Big Data While Shaking Up the Peer Review Process -- 1. Introduction -- 2. The Crowd Peer Review Process -- 2.1 Reviewer's Feedback -- 2.2 Conclusions -- 3. Meta-Analysis in Biocomputing -- 3.1 Novel Methods for Meta-Analysis of 'Omics Data -- 3.2 Using Publicly Available Data in Methods Development -- 3.3 Studying the Structure of Publicly Available Data -- 3.4 Conclusions -- Acknowledgements -- References -- Multitask Group Lasso for Genome Wide Association Studies in Diverse Populations -- 1. Introduction -- 2. Methods -- 2.1. Population stratification -- 2.2.1. Adjacency-constrained hierarchical clustering -- 2.2.2. LD-groups across populations -- 2.3. Multitask group Lasso. , 2.3.1. General framework and problem formulation -- 2.3.2. Related work -- 2.3.3. Gap safe screening rules -- 2.4. Stability selection -- 3. Experiments -- 3.1. Data -- 3.2. Preprocessing -- 3.3. Comparison partners -- 4. Results -- 4.1. MuGLasso draws on both LD-groups and the multitask approach to recover disease SNPs -- 4.2. MuGLasso provides the most stable selection -- 4.3. MuGLasso selects both task-speci c and global LD-groups -- 5. Discussion and Conclusions -- Acknowledgments -- Supplementary Materials and code -- References -- Mixed Effects Machine Learning Models for Colon Cancer Metastasis Prediction Using Spatially Localized Immuno-Oncology Markers -- 1. Introduction -- 2. Motivation for Comparison Study -- 2.1. Review of Prior Spatial Omics Analysis Methods -- 2.2. Motivation for Mixed Effects Machine Learning Approaches -- 3. Materials and Methods -- 3.1. Data Acquisition and Preprocessing -- 3.2. Experimental Design: Prediction Tasks and Modeling Approaches -- 4. Results -- 4.1. Macro: Inter-Tumoral Prediction -- 4.2. METS: Nodal and Distant Metastasis Prediction -- 5. Discussion -- 6. Conclusion -- 7. Acknowledgements -- 8. References -- Improving QSAR Modeling for Predictive Toxicology Using Publicly Aggregated Semantic Graph Data and Graph Neural Networks -- 1. Introduction -- 2. Methods -- 2.1. Obtaining toxicology assay data -- 2.2. Aggregating publicly available multimodal graph data -- 2.3. Heterogeneous graph neural network -- 2.3.1. Node classification -- 2.4. Baseline QSAR classifiers -- 3. Results -- 3.1. GNN node classification performance vs. baseline QSAR models -- 3.2. Ablation analysis of graph components' inuence on the trained model -- 4. Discussion -- 4.1. GNNs versus traditional ML for QSAR modeling -- 4.2. Interpretability of GNNs in QSAR -- 4.3. Sources of bias and their effects on QSAR for toxicity prediction. , 5. Conclusions.

Additional Edition: ISBN 981-12-5046-4

Additional Edition: ISBN 981-12-5047-2

Language: English

UID:

Format: 1 online resource (431 pages)

ISBN: 9789811250477

Note: Intro -- Content -- Preface -- AI-DRIVEN ADVANCES IN MODELING OF PROTEIN STRUCTURE -- Session Introduction: AI-Driven Advances in Modeling of Protein Structure -- 1. A short retrospect -- 2. A brief outline of current research -- 3. Future developments (complexes, ligand interactions, other molecules, dynamics, language models, geometry models, sequence design) -- 4. What is needed for further progress? -- 5. Overview of papers in this session -- 5.1. Evaluating significance of training data selection in machine learning -- 5.2. Geometric pattern transferability -- 5.3. Supervised versus unsupervised sequence to contact learning -- 5.4. Side chain packing using SE(3) transformers -- 5.5. Feature selection in electrostatic representations of ligand binding sites -- References -- Training Data Composition Affects Performance of Protein Structure Analysis Algorithms -- 1. Introduction -- 2. Methods -- 2.1. Experimental Design -- 2.2. Task-specific Methods -- 3. Results -- 3.1. Performance on NMR and cryo-EM structures is consistently lower than performance on X-ray structures, independent of training set -- 3.2. Inclusion of NMR data in the training set improves performance on held-out NMR data and does not degrade performance on X-ray data -- 3.3. Known biochemical and biophysical effects are replicated in trained models -- 3.4. Downsampling X-ray structures during training negatively affects performance on all types of data -- 4. Conclusion -- 5. Acknowledgments -- References -- Transferability of Geometric Patterns from Protein Self-Interactions to Protein-Ligand Interactions -- 1. Introduction -- 2. Related Work -- 3. Methods -- 3.1. Datasets -- 3.2. Contact extraction -- 3.3. Representing contact geometry -- 4. Results -- 4.1. Protein self-contacts exhibit clear geometric clustering. , 4.2. Many geometric patterns transfer to protein-ligand contacts -- 4.3. Application to protein-ligand docking -- 5. Conclusion and Future Work -- Supplemental Material, Code, and Data Availability -- Acknowledgments -- References -- Interpreting Potts and Transformer Protein Models Through the Lens of Simplified Attention -- 1. Introduction -- 2. Background -- 3. Methods -- 3.1. Potts Models -- 3.2. Factored Attention -- 3.3. Single-layer attention -- 3.4. Pretraining on Sequence Databases -- 3.5. Extracting Contacts -- 4. Results -- 5. Discussion -- Acknowledgements -- References -- Side-Chain Packing Using SE(3)-Transformer -- 1. Introduction -- 2. Methods -- 2.1. Neighborhood Graph Representation -- 2.2. The SE(3)-Transformer Architecture -- 2.3. Node Features -- 2.4. Final Layer -- 2.5. Rotamer Selection -- 2.6. Experiments -- 3. Results -- 4. Conclusion -- 5. Acknowledgements -- 6. References -- DeepVASP-E: A Flexible Analysis of Electrostatic Isopotentials for Finding and Explaining Mechanisms that Control Binding Specificity -- 1. Introduction -- 2. Methods -- 2.1. Convolutional Neural Network -- 2.2. Experimental Design -- 2.3. Comparison with Existing Methods -- 3. Results -- 4. Conclusions -- Acknowledgements -- References -- BIG DATA IMAGING GENOMICS -- Session Introduction: Big Data Imaging Genomics -- 1. Introduction -- 2. Overview of Contributions -- References -- A New Mendelian Randomization Method to Estimate Causal Effects of Multivariable Brain Imaging Exposures -- 1. Introduction -- 2. Methods -- 2.1. Step 1 : Mendelian randomization analysis on a single imaging exposure -- 2.2. Step 2: Joint instrumental variables and imaging exposures selection -- 2.3. Step 3: Causal effect identification for multiple imaging exposures -- 3. Application to evaluate the causal effect of white matter microstructure integrity on cognitive function. , 3.1. Data and study cohort -- 3.2. Results -- 4. Simulation -- 5. Discussion -- Funding -- Availability of data and materials -- Authors' contributions -- References -- Efficient Differentially Private Methods for a Transmission Disequilibrium Test in Genome Wide Association Studies -- 1. Introduction -- 2. Preliminaries -- 2.1. TDT -- 2.2. Differential Privacy -- 3. Methods -- 3.1. Exact Algorithm -- 3.2. Approximation Algorithm -- 4. Experiments -- 4.1. Simulation Data -- 4.2. Results -- 4.2.1. Run Time -- 4.2.2. Accuracy -- 4.3. Real Data -- 5. Conclusion -- Acknowledgement -- References -- Identifying Imaging Genetic Associations via Regional Morphometricity Estimation -- 1. Introduction -- 2. Methods -- 3. Materials -- 4. Experimental Design -- 5. Results and Discussion -- 6. Conclusion -- Acknowledgements -- References -- Identifying Highly Heritable Brain Amyloid Phenotypes Through Mining Alzheimer's Imaging and Sequencing Biobank Data -- 1. Introduction -- 2. Method -- 3. Materials -- 4. Experimental Workow -- 5. Results and Discussion -- 6. Conclusion -- Acknowledgements -- References -- Effects of ApoE4 and ApoE2 Genotypes on Subcortical Magnetic Susceptibility and Microstructure in 27,535 Participants from the UK Biobank -- 1. Introduction -- 2. Methods -- 2.1. UK Biobank Participants -- 2.2. T1-Weighted MRI -- 2.3. Quantitative Magnetic Susceptibility -- 2.4. Diffusion-Weighted MRI -- 2.5. Statistical Analyses -- 3. Results -- 3.1. ApoE4 Microstructural Associations -- 3.2. ApoE2 Microstructural Associations -- 3.3. ApoE-by-Age Interactions -- 3.3.1. ApoE Associations Stratified by Age -- 4. Discussion -- References -- Separating Clinical and Subclinical Depression by Big Data Informed Structural Vulnerability Index and Its impact on Cognition: ENIGMA Dot Product -- 1. Introduction -- 2. Methods. -- 2.1 Participants. , 2.2 Major Depressive Disorder Classification -- 2.3 Imaging Protocol and Processing -- 2.4 Calculation of linear indices of similarity -- 2.5 Calculation of QRI -- 2.7 Cognitive assessment -- 2.8 Statistics -- 3. Results -- 3.1 Group differences in symptoms and biomarkers -- 3.2 Effects of MDD on cognition. -- 3.3. Cognitive association -- 4. Discussion. -- 5. Conclusion -- 6. Acknowledgement -- References -- Generalizing Few-Shot Classification of Whole-Genome Doubling Across Cancer Types -- 1. Introduction -- 2. Related Work -- 3. Cohort -- 3.1. Cohort Selection -- 3.2. Feature Extraction -- 4. Methods -- 4.1. Model -- 4.2. Training -- 4.2.1. Pre-Training -- 4.2.2. Meta-Training -- 4.3. Meta-Validation and Meta-Test -- 4.4. Experiments -- 4.4.1. Cancer Types -- 4.4.2. Batch Effects -- 5. Results -- 5.1. Cancer Types -- 5.2. Batch Effects -- 5.2.1. Image Resolution -- 5.2.2. Image Brightness -- 6. Discussion -- Software and Data -- References -- HUMAN INTRIGUE: META-ANALYSIS APPROACHES FOR BIG QUESTIONS WITH BIG DATA WHILE SHAKING UP THE PEER REVIEW PROCESS -- Session Introduction: Human Intrigue: Meta-Analysis Approaches for Big Questions with Big Data While Shaking Up the Peer Review Process -- 1. Introduction -- 2. The Crowd Peer Review Process -- 2.1 Reviewer's Feedback -- 2.2 Conclusions -- 3. Meta-Analysis in Biocomputing -- 3.1 Novel Methods for Meta-Analysis of 'Omics Data -- 3.2 Using Publicly Available Data in Methods Development -- 3.3 Studying the Structure of Publicly Available Data -- 3.4 Conclusions -- Acknowledgements -- References -- Multitask Group Lasso for Genome Wide Association Studies in Diverse Populations -- 1. Introduction -- 2. Methods -- 2.1. Population stratification -- 2.2.1. Adjacency-constrained hierarchical clustering -- 2.2.2. LD-groups across populations -- 2.3. Multitask group Lasso. , 2.3.1. General framework and problem formulation -- 2.3.2. Related work -- 2.3.3. Gap safe screening rules -- 2.4. Stability selection -- 3. Experiments -- 3.1. Data -- 3.2. Preprocessing -- 3.3. Comparison partners -- 4. Results -- 4.1. MuGLasso draws on both LD-groups and the multitask approach to recover disease SNPs -- 4.2. MuGLasso provides the most stable selection -- 4.3. MuGLasso selects both task-speci c and global LD-groups -- 5. Discussion and Conclusions -- Acknowledgments -- Supplementary Materials and code -- References -- Mixed Effects Machine Learning Models for Colon Cancer Metastasis Prediction Using Spatially Localized Immuno-Oncology Markers -- 1. Introduction -- 2. Motivation for Comparison Study -- 2.1. Review of Prior Spatial Omics Analysis Methods -- 2.2. Motivation for Mixed Effects Machine Learning Approaches -- 3. Materials and Methods -- 3.1. Data Acquisition and Preprocessing -- 3.2. Experimental Design: Prediction Tasks and Modeling Approaches -- 4. Results -- 4.1. Macro: Inter-Tumoral Prediction -- 4.2. METS: Nodal and Distant Metastasis Prediction -- 5. Discussion -- 6. Conclusion -- 7. Acknowledgements -- 8. References -- Improving QSAR Modeling for Predictive Toxicology Using Publicly Aggregated Semantic Graph Data and Graph Neural Networks -- 1. Introduction -- 2. Methods -- 2.1. Obtaining toxicology assay data -- 2.2. Aggregating publicly available multimodal graph data -- 2.3. Heterogeneous graph neural network -- 2.3.1. Node classification -- 2.4. Baseline QSAR classifiers -- 3. Results -- 3.1. GNN node classification performance vs. baseline QSAR models -- 3.2. Ablation analysis of graph components' inuence on the trained model -- 4. Discussion -- 4.1. GNNs versus traditional ML for QSAR modeling -- 4.2. Interpretability of GNNs in QSAR -- 4.3. Sources of bias and their effects on QSAR for toxicity prediction. , 5. Conclusions.

Additional Edition: Print version: Altman, Russ B Biocomputing 2022 - Proceedings Of The Pacific Symposium Singapore : World Scientific Publishing Company,c2021 ISBN 9789811250460

Language: English

Keywords: Electronic books.

URL: Click to View

UID:

Format: 1 online resource (431 pages)

Note: Intro -- Content -- Preface -- AI-DRIVEN ADVANCES IN MODELING OF PROTEIN STRUCTURE -- Session Introduction: AI-Driven Advances in Modeling of Protein Structure -- 1. A short retrospect -- 2. A brief outline of current research -- 3. Future developments (complexes, ligand interactions, other molecules, dynamics, language models, geometry models, sequence design) -- 4. What is needed for further progress? -- 5. Overview of papers in this session -- 5.1. Evaluating significance of training data selection in machine learning -- 5.2. Geometric pattern transferability -- 5.3. Supervised versus unsupervised sequence to contact learning -- 5.4. Side chain packing using SE(3) transformers -- 5.5. Feature selection in electrostatic representations of ligand binding sites -- References -- Training Data Composition Affects Performance of Protein Structure Analysis Algorithms -- 1. Introduction -- 2. Methods -- 2.1. Experimental Design -- 2.2. Task-specific Methods -- 3. Results -- 3.1. Performance on NMR and cryo-EM structures is consistently lower than performance on X-ray structures, independent of training set -- 3.2. Inclusion of NMR data in the training set improves performance on held-out NMR data and does not degrade performance on X-ray data -- 3.3. Known biochemical and biophysical effects are replicated in trained models -- 3.4. Downsampling X-ray structures during training negatively affects performance on all types of data -- 4. Conclusion -- 5. Acknowledgments -- References -- Transferability of Geometric Patterns from Protein Self-Interactions to Protein-Ligand Interactions -- 1. Introduction -- 2. Related Work -- 3. Methods -- 3.1. Datasets -- 3.2. Contact extraction -- 3.3. Representing contact geometry -- 4. Results -- 4.1. Protein self-contacts exhibit clear geometric clustering. , 4.2. Many geometric patterns transfer to protein-ligand contacts -- 4.3. Application to protein-ligand docking -- 5. Conclusion and Future Work -- Supplemental Material, Code, and Data Availability -- Acknowledgments -- References -- Interpreting Potts and Transformer Protein Models Through the Lens of Simplified Attention -- 1. Introduction -- 2. Background -- 3. Methods -- 3.1. Potts Models -- 3.2. Factored Attention -- 3.3. Single-layer attention -- 3.4. Pretraining on Sequence Databases -- 3.5. Extracting Contacts -- 4. Results -- 5. Discussion -- Acknowledgements -- References -- Side-Chain Packing Using SE(3)-Transformer -- 1. Introduction -- 2. Methods -- 2.1. Neighborhood Graph Representation -- 2.2. The SE(3)-Transformer Architecture -- 2.3. Node Features -- 2.4. Final Layer -- 2.5. Rotamer Selection -- 2.6. Experiments -- 3. Results -- 4. Conclusion -- 5. Acknowledgements -- 6. References -- DeepVASP-E: A Flexible Analysis of Electrostatic Isopotentials for Finding and Explaining Mechanisms that Control Binding Specificity -- 1. Introduction -- 2. Methods -- 2.1. Convolutional Neural Network -- 2.2. Experimental Design -- 2.3. Comparison with Existing Methods -- 3. Results -- 4. Conclusions -- Acknowledgements -- References -- BIG DATA IMAGING GENOMICS -- Session Introduction: Big Data Imaging Genomics -- 1. Introduction -- 2. Overview of Contributions -- References -- A New Mendelian Randomization Method to Estimate Causal Effects of Multivariable Brain Imaging Exposures -- 1. Introduction -- 2. Methods -- 2.1. Step 1 : Mendelian randomization analysis on a single imaging exposure -- 2.2. Step 2: Joint instrumental variables and imaging exposures selection -- 2.3. Step 3: Causal effect identification for multiple imaging exposures -- 3. Application to evaluate the causal effect of white matter microstructure integrity on cognitive function. , 3.1. Data and study cohort -- 3.2. Results -- 4. Simulation -- 5. Discussion -- Funding -- Availability of data and materials -- Authors' contributions -- References -- Efficient Differentially Private Methods for a Transmission Disequilibrium Test in Genome Wide Association Studies -- 1. Introduction -- 2. Preliminaries -- 2.1. TDT -- 2.2. Differential Privacy -- 3. Methods -- 3.1. Exact Algorithm -- 3.2. Approximation Algorithm -- 4. Experiments -- 4.1. Simulation Data -- 4.2. Results -- 4.2.1. Run Time -- 4.2.2. Accuracy -- 4.3. Real Data -- 5. Conclusion -- Acknowledgement -- References -- Identifying Imaging Genetic Associations via Regional Morphometricity Estimation -- 1. Introduction -- 2. Methods -- 3. Materials -- 4. Experimental Design -- 5. Results and Discussion -- 6. Conclusion -- Acknowledgements -- References -- Identifying Highly Heritable Brain Amyloid Phenotypes Through Mining Alzheimer's Imaging and Sequencing Biobank Data -- 1. Introduction -- 2. Method -- 3. Materials -- 4. Experimental Workow -- 5. Results and Discussion -- 6. Conclusion -- Acknowledgements -- References -- Effects of ApoE4 and ApoE2 Genotypes on Subcortical Magnetic Susceptibility and Microstructure in 27,535 Participants from the UK Biobank -- 1. Introduction -- 2. Methods -- 2.1. UK Biobank Participants -- 2.2. T1-Weighted MRI -- 2.3. Quantitative Magnetic Susceptibility -- 2.4. Diffusion-Weighted MRI -- 2.5. Statistical Analyses -- 3. Results -- 3.1. ApoE4 Microstructural Associations -- 3.2. ApoE2 Microstructural Associations -- 3.3. ApoE-by-Age Interactions -- 3.3.1. ApoE Associations Stratified by Age -- 4. Discussion -- References -- Separating Clinical and Subclinical Depression by Big Data Informed Structural Vulnerability Index and Its impact on Cognition: ENIGMA Dot Product -- 1. Introduction -- 2. Methods. -- 2.1 Participants. , 2.2 Major Depressive Disorder Classification -- 2.3 Imaging Protocol and Processing -- 2.4 Calculation of linear indices of similarity -- 2.5 Calculation of QRI -- 2.7 Cognitive assessment -- 2.8 Statistics -- 3. Results -- 3.1 Group differences in symptoms and biomarkers -- 3.2 Effects of MDD on cognition. -- 3.3. Cognitive association -- 4. Discussion. -- 5. Conclusion -- 6. Acknowledgement -- References -- Generalizing Few-Shot Classification of Whole-Genome Doubling Across Cancer Types -- 1. Introduction -- 2. Related Work -- 3. Cohort -- 3.1. Cohort Selection -- 3.2. Feature Extraction -- 4. Methods -- 4.1. Model -- 4.2. Training -- 4.2.1. Pre-Training -- 4.2.2. Meta-Training -- 4.3. Meta-Validation and Meta-Test -- 4.4. Experiments -- 4.4.1. Cancer Types -- 4.4.2. Batch Effects -- 5. Results -- 5.1. Cancer Types -- 5.2. Batch Effects -- 5.2.1. Image Resolution -- 5.2.2. Image Brightness -- 6. Discussion -- Software and Data -- References -- HUMAN INTRIGUE: META-ANALYSIS APPROACHES FOR BIG QUESTIONS WITH BIG DATA WHILE SHAKING UP THE PEER REVIEW PROCESS -- Session Introduction: Human Intrigue: Meta-Analysis Approaches for Big Questions with Big Data While Shaking Up the Peer Review Process -- 1. Introduction -- 2. The Crowd Peer Review Process -- 2.1 Reviewer's Feedback -- 2.2 Conclusions -- 3. Meta-Analysis in Biocomputing -- 3.1 Novel Methods for Meta-Analysis of 'Omics Data -- 3.2 Using Publicly Available Data in Methods Development -- 3.3 Studying the Structure of Publicly Available Data -- 3.4 Conclusions -- Acknowledgements -- References -- Multitask Group Lasso for Genome Wide Association Studies in Diverse Populations -- 1. Introduction -- 2. Methods -- 2.1. Population stratification -- 2.2.1. Adjacency-constrained hierarchical clustering -- 2.2.2. LD-groups across populations -- 2.3. Multitask group Lasso. , 2.3.1. General framework and problem formulation -- 2.3.2. Related work -- 2.3.3. Gap safe screening rules -- 2.4. Stability selection -- 3. Experiments -- 3.1. Data -- 3.2. Preprocessing -- 3.3. Comparison partners -- 4. Results -- 4.1. MuGLasso draws on both LD-groups and the multitask approach to recover disease SNPs -- 4.2. MuGLasso provides the most stable selection -- 4.3. MuGLasso selects both task-speci c and global LD-groups -- 5. Discussion and Conclusions -- Acknowledgments -- Supplementary Materials and code -- References -- Mixed Effects Machine Learning Models for Colon Cancer Metastasis Prediction Using Spatially Localized Immuno-Oncology Markers -- 1. Introduction -- 2. Motivation for Comparison Study -- 2.1. Review of Prior Spatial Omics Analysis Methods -- 2.2. Motivation for Mixed Effects Machine Learning Approaches -- 3. Materials and Methods -- 3.1. Data Acquisition and Preprocessing -- 3.2. Experimental Design: Prediction Tasks and Modeling Approaches -- 4. Results -- 4.1. Macro: Inter-Tumoral Prediction -- 4.2. METS: Nodal and Distant Metastasis Prediction -- 5. Discussion -- 6. Conclusion -- 7. Acknowledgements -- 8. References -- Improving QSAR Modeling for Predictive Toxicology Using Publicly Aggregated Semantic Graph Data and Graph Neural Networks -- 1. Introduction -- 2. Methods -- 2.1. Obtaining toxicology assay data -- 2.2. Aggregating publicly available multimodal graph data -- 2.3. Heterogeneous graph neural network -- 2.3.1. Node classification -- 2.4. Baseline QSAR classifiers -- 3. Results -- 3.1. GNN node classification performance vs. baseline QSAR models -- 3.2. Ablation analysis of graph components' inuence on the trained model -- 4. Discussion -- 4.1. GNNs versus traditional ML for QSAR modeling -- 4.2. Interpretability of GNNs in QSAR -- 4.3. Sources of bias and their effects on QSAR for toxicity prediction. , 5. Conclusions.

Additional Edition: ISBN 981-12-5046-4

Additional Edition: ISBN 981-12-5047-2

Language: English