Deep Learning in Genetics and Genomics

Volume 1: Foundations and Introductory Applications

1st Edition - November 28, 2024
Editor: Khalid Raza
Language: English
Paperback ISBN:
9 7 8 - 0 - 4 4 3 - 2 7 5 7 4 - 6
eBook ISBN:
9 7 8 - 0 - 4 4 3 - 2 7 5 7 5 - 3

Deep Learning in Genetics and Genomics vol. 1, Foundations and Applications, the intersection of deep learning and genetics opens up new avenues for advancing our unders… Read more

Purchase options

Institutional subscription on ScienceDirect

Request a sales quote

Deep Learning in Genetics and Genomics vol. 1, Foundations and Applications, the intersection of deep learning and genetics opens up new avenues for advancing our understanding of the genetic code, gene regulation, and the broader genomics landscape. The book not only covers the most up-to-date advancements in the field of deep learning in genetics and genomics, but also a wide spectrum of (sub) topics including medical and clinical genetics, predictive medicine, transcriptomic, and gene expression studies. In 21 chapters Deep Learning in Genetics and Genomics vol. 1, Foundations and Applications describes how AI and DL have become increasingly useful in genetics and genomics research where both play a crucial role by accelerating research, improving the understanding of the human genome, and enabling personalized healthcare. From the fundamentals concepts and practical applications of deep learning algorithms to a wide range of challenging problems from genetics and genomics, Deep Learning in Genetics and Genomics vol. 1, Foundations and Applications creates a better knowledge of the biological and genetics mechanisms behind disease illnesses and improves the forecasting abilities using the different methodologies described. This title offers a unique resource for wider, deeper, and in-depth coverage of recent advancement in deep learning-based approaches in genetics and genomics, helping researchers process and interpret vast amounts of genetic data, identify patterns, and make discoveries that would be challenging or impossible using traditional methods.

Title of Book
Cover image
Title page
Table of Contents
Copyright
Dedication
Contributors
About the editor
Preface
Acknowledgments
About the book
Chapter 1. Basics of genetics and genomics
1 Introduction
1.1 Structure of DNA
1.2 Central dogma of biology
1.3 DNA replication
1.4 DNA transcription and protein synthesis
2 Structure of genes and genomes
2.1 Prokaryote genome structure
2.2 Eukaryote genome structure
3 Era of genetics
3.1 The monohybrid cross
3.2 The dihybrid cross
3.3 Deviations from Mendel's laws
4 Era of genomics
4.1 Polymerase chain reaction (PCR)
4.2 Early genomic experiments
4.3 DNA sequencing
4.4 High-throughput sequencing
4.5 Next generation sequencing
4.5.1 Pyrosequencing
4.5.2 Ion Torrent sequencing
4.5.3 Illumina sequencing by synthesis
4.6 Third generation sequencing
4.6.1 Oxford Nanopore Technologies sequencing
4.6.2 PacBio sequencing
5 Data analysis of genomic experiments—An overview
6 Types of genomics experiments
6.1 Structural genomics
6.2 Functional genomics
6.3 Applications of genomics
7 Postgenomic era
8 Conclusion
Chapter 2. Introduction to deep learning for genomics
1 Introduction
1.1 Specific contributions
1.2 Background
2 Fundamentals of deep learning
2.1 Basic concepts
2.2 Deep learning architectures
2.3 Convolutional neural networks
2.4 Feedforward neural networks
2.5 Recurrent neural networks
2.6 Long short-term memory networks
2.7 Mathematical description of LSTM
2.8 Gated recurrent units
2.9 Generative adversarial networks
3 Applications of deep learning in genomics
3.1 Sequence analysis
3.2 Gene expression prediction
3.3 Disease prediction and personalized medicine
4 Advanced deep learning techniques in genomics
4.1 Attention mechanisms and transformer models
4.2 Generative adversarial networks
4.3 Integration with emerging technologies
5 Disadvantages and flaws of past machine learning technologies
5.1 Key limitations
5.2 Case study: GPT-3
5.3 Improvements in GPT-4
6 Challenges and solutions in deep learning for genomics
6.1 Data challenges
6.2 Model interpretability and explainability
6.3 Computational and resource constraints
7 Ethical and societal implications
7.1 Ethical considerations
7.2 Societal impact
8 Future directions and emerging trends
8.1 Trends in deep learning for genomics
8.2 Research directions
9 Conclusion
Chapter 3. Foundations and applications of computational genomics
1 Introduction
2 Computational genomics and human genomes
3 Computational tools and techniques
4 Computational genomics and genetic diseases
5 Computation genomics in application of analysis genetic diseases
6 Case study on genetic disease analysis
7 Limitations and future directions
8 Conclusions
Chapter 4. Decoding DNA: Deep learning's impact on genomic exploration
1 Introduction
2 Deep learning methodologies for decoding DNA
2.1 Neural networks and genomic pattern recognition
2.2 Processing and analysing large genomic datasets
2.3 Case studies of novel deep learning applications
3 Unlocking the secrets of the genome
3.1 Advances in gene sequencing analysis and insight
3.2 Identifying critical sequences and genetic variants
3.3 Identifying intricate correlations and relationships within DNA
3.4 Examples of recent breakthrough discoveries
4 Improving prediction and diagnosis of disease
4.1 Predicting disease risks from genomic data
4.2 Enabling earlier and more accurate diagnoses
4.3 Potential for precision and personalized medicine
4.4 Promises and current limitations
4.4.1 Promises
4.4.2 Current limitations
4.5 Future horizons: The path ahead for AI and genomics
5 Conclusion
Chapter 5. AI and deep learning in cancer genomics
1 Introduction
2 Application of AI to cancer genomics data
3 Machine learning for the classification of different cancer by using gene expression data
4 Techniques of deep learning in the prognosis of cancer with genomics data
5 Genomics-based AI in immunotherapy prediction
6 AI and ML in genomic and precision medicine
7 Challenges
7.1 Technical challenges
7.2 Ethical challenges
8 Future directions
9 Conclusion
List of abbreviations
Chapter 6. Unravelling the recent developments in applications and challenges of AI in cancer biology: An overview
1 Introduction
2 Prognosis
3 Diagnosis
4 Cancer treatment and management
4.1 AI and chemotherapy, radiotherapy and immunotherapy
4.2 Role of AI in cancer overtreatment and clinical decision support systems
5 Hurdles for real-life deployment of AI
5.1 Data, data privacy, and protection-related challenges
6 Conclusion
Chapter 7. Unlocking the potential of deep learning for oncological sequence analysis: A review
1 Introduction
2 Essentials of oncological sequences
2.1 Types of oncological sequences
2.2 Challenges in oncological sequence analysis
3 Overview of deep learning
3.1 Deep learning architectures
3.2 Convolutional neural networks
3.3 Recurrent neural networks
3.4 Long short-term memory networks
4 Deep learning applications in oncological sequence analysis
4.1 Deep learning in genomics
4.2 Deep learning in transcriptomics
4.3 Deep learning in proteomics
5 Challenges and future perspectives
6 Conclusion
Chapter 8. Deep learning in medical genetics: A review
1 Introduction
2 Datasets
2.1 Genomic benchmarks
2.2 Genomics data lake
2.3 Genomic data analysis
3 Deep learning approaches
3.1 Supervised deep learning methods
3.2 Unsupervised deep learning methods
3.3 Semi-supervised deep learning methods
4 Future scope and challenges
5 Conclusion
Chapter 9. Navigating the genomic landscape: A deep dive into clinical genetics with deep learning
1 Introduction
2 Deep learning
2.1 Supervised learning
2.2 Unsupervised learning
2.3 Semi-supervised learning
3 Deep learning tools in genomics
4 Crucial role of deep learning in genomics
4.1 Deep learning at DNA level
4.1.1 Promoter
4.1.2 Enhancer
4.1.3 Noncoding region
4.1.4 Interactions between genomic elements
4.1.5 Other domains
4.2 Deep learning at RNA level
4.2.1 Splicing
4.2.2 Noncoding RNA
4.2.3 Messenger RNA
4.2.4 Expression
4.3 Deep learning at protein level
4.3.1 Transcription factor
4.3.2 RNA-specific binding proteins
5 Deep learning in surgery
6 Promising approach of deep learning in the detection of genetic disorders
6.1 Neurology
6.2 Fetal ultrasound
6.3 Polycystic ovary syndrome
6.4 Epilepsy
6.5 Cardiovascular diseases
7 Implications of deep learning in the medical health computational field
7.1 Medical image
7.2 Electronic health record
7.3 Drug discovery
7.4 Oncology
8 Deep learning application in longitudinal datasets
9 Challenges of deep learning with their alternate solutions
9.1 Training data
9.2 Interpretability of data
9.3 Uncertainty scaling
10 Conclusion and future perspective
Chapter 10. Advancing clinical genomics: Bridging the gap between deep learning models and interpretability for improved decision support
1 Introduction
1.1 Background on clinical genomics
1.2 The role of deep learning models in clinical genomics
1.3 Challenges limiting adoption
1.4 The critical need for interpretability and explainability
1.5 Purpose of the study
2 Overview of deep learning models in clinical genomics
2.1 Current state of interpretability in deep learning models
2.2 Existing challenges and limitations
2.3 Recent advances in bridging the gap
3 Methodology
3.1 Search strategy
3.2 Inclusion and exclusion criteria
4 Techniques for improved interpretability
4.1 Overview of proposed methodologies
4.2 Detailed explanation of novel techniques
4.3 Comparative analysis with existing approaches
4.4 Evaluation metrics for interpretability
5 Application of interpretability techniques in clinical settings
5.1 Case studies and real-world applications
5.2 Impact on healthcare professionals and clinical decision making
5.3 Integration into routine clinical workflows
5.4 Challenges and considerations for implementation
6 Future directions and implications
6.1 Potential benefits of enhanced interpretability
6.2 Areas for further research and development
6.3 Ethical and regulatory considerations
6.4 Implications for advancing precision medicine
7 Conclusion
7.1 Summary of key findings
7.2 Contributions to the field of clinical genomics
7.3 Recommendations for future practice and policy
Chapter 11. Deep learning in clinical genomics-based cancer diagnosis
1 Introduction
2 Advances in deep learning in clinical sciences
2.1 Clinical imaging
2.2 Disease diagnosis
2.3 Drug discovery and development
2.4 Health monitoring and overall management
2.5 Personalized medicine
3 AI methods in clinical laboratory
3.1 Pan-omics approaches in healthcare
3.1.1 Precision oncology
3.1.2 Cardiovascular disease risk prediction
3.1.3 Neurodegenerative disease research
3.1.4 Role of genome (or other technology for ex RNA Seq, etc.)
3.1.5 Role in next-generation sequencing
3.1.6 Role in whole genome and exome sequencing
3.1.7 Role in variant discovery
3.1.8 Role in healthcare systems
4 Impact of deep learning in clinical genomics for cancer
5 Impact of deep learning in clinical genomics for disease classification
6 Current challenges and prospects
7 Conclusion
Chapter 12. Deep learning in predictive medicine: Current state of the art
1 Introduction
1.1 Overview of predictive medicine
1.2 Understanding deep learning in healthcare
1.3 Significance and applications
2 Data acquisition and preprocessing for predictive analysis
2.1 Medical data resources
2.1.1 Electronic health record
2.1.2 Administrative data resources
2.1.3 Patient disease registries
2.1.4 Disease specific data resources
2.1.5 Clinical trials databases
2.2 Data collection challenges and solutions
2.2.1 Data quality and consistency
2.2.2 Data security and privacy
2.2.3 Data integration and interoperability
2.2.4 Data analysis and visualization
2.2.5 Data ethics and governance
2.2.6 Data literacy and culture
2.3 Data preprocessing techniques
2.3.1 Data cleaning
2.3.2 Data integration
2.3.3 Data transformation
2.3.4 Feature engineering and selection
3 Basics of deep learning in healthcare
3.1 Neural networks and deep learning architectures
3.1.1 Architecture of a single artificial neuron
3.1.2 Architecture of multilayered neural network
3.2 Deep learning algorithms in medical applications
3.2.1 The perceptron
3.2.2 Feed forward neural network
3.2.3 Residual networks
3.2.4 Recurrent neural networks
3.2.5 The long short-term memory network
3.2.6 Convolution neural network
3.2.7 Other neural networks
4 Predictive modeling in early disease diagnosis using deep learning
4.1 Deep learning in early diagnosis of cancer
4.2 Deep learning in diabetic retinopathy
4.3 Deep learning in cardiovascular disorders
4.4 Deep learning in neurological disorders
5 Deep learning applications in treatment and patient care
5.1 Personalized medicine and treatment plans
6 Ethical and regulatory considerations in predictive medicine
6.1 Privacy and security concerns
6.2 Compliance with healthcare regulations
6.3 Ethical implications and best practices
7 Current challenges and future directions
7.1 Current challenges in implementing deep learning in healthcare
7.2 Future trends and opportunities
7.3 Unexplored areas and potential developments
8 Conclusion
Chapter 13. Applications of AI in cancer genomics: A way toward intelligent decision systems in healthcare
1 Introduction
2 Current status of AI technology
3 AI in medicine: The new road
4 Pan-omics in clinical healthcare
4.1 Genomics
4.2 Epigenomics
4.3 Transcriptomics
4.4 Proteomics
4.5 Metabolomics
5 Pan-omics approaches used in healthcare
6 Applications of AI in medicine
7 Application of AI in clinical cancer genomics: A way toward intelligent systems
8 Conclusion
8.1 Take-home message
Chapter 14. The role of deep learning in drug discovery
1 Introduction
2 Deep learning techniques
2.1 Classic neural networks
2.2 Convolutional neural networks
2.3 Recurrent neural networks
2.4 Hybrid architectures
3 Deep learning applications in drug discovery problems
3.1 Drug–target interaction prediction using DL
3.1.1 Drug-based models
3.1.2 Structure (graph)-based models
3.1.3 Drug–protein (disease)-based models
4 Drug sensitivity and response prediction using DL
5 Conclusion
Chapter 15. Deep learning applications in genomics-based toxicology assessment
1 Introduction
1.1 Genomics-based toxicology assessment
1.2 Deep learning: unleashing the power of machine learning
2 Overview of deep learning methods
2.1 Convolutional neural networks for genomics sequences
2.2 Recurrent neural networks for time-series data
3 Generative adversarial networks for genomics data augmentation
4 Real-world applications of deep learning in genomics-based toxicology and the used computational platforms
4.1 Case study on machine learning based drug-induced genotoxicity prediction model by evaluating physiochemical descriptors: AI approach
5 Challenges and opportunities: Navigating the complex landscape
6 Ethical considerations
7 Future scopes and directions
8 Limitation of deep learning in toxicology
9 Conclusion
Chapter 16. The revolutionary impact of deep learning in transcriptomics and gene expression analysis: A genomic paradigm shift
1 Introduction
1.1 Objectives
2 Related study
3 Problem statement
4 Methodology
4.1 Algorithm
4.2 Mathematical formalization
5 Result analysis
6 Simulation parameter
7 Conclusion
8 Future scope
Chapter 17. Data-driven genomics: A triad of big data, cloud, and IoT in genomics research
1 Introduction
2 Big data in genomics research
2.1 Challenges and opportunities
2.2 Applications
3 Cloud computing in genomics research
3.1 Challenges and opportunities
3.2 Applications
4 Internet of Things in genomics research
4.1 Challenges
4.2 Applications
5 Genomic data safety and security
5.1 Privacy concerns
5.2 Informed consent
5.3 Data security
5.4 Data sharing
6 How is triad accelerating genomic research?
6.1 Facilitating large-scale data acquisition and storage
6.2 Enabling powerful data analysis and processing
6.3 Fostering collaboration and data sharing
6.4 Some examples and cases
7 Case studies and future directions
7.1 Case studies
7.1.1 Case #1: All of Us Research Program
7.1.2 Case #2: Automation of genomic data analysis using IoT and Eoulsan
7.1.3 Case #3: Rainbow: Addressing challenges of big data in cloud computing and genomics
7.2 Future directions
8 Conclusion
Chapter 18. Deep learning in variant detection and annotation
1 Introduction
2 Fundamentals to advances in variant detection
2.1 Basics of genetic variants
2.2 Types of genetic variants
2.2.1 Single-nucleotide polymorphisms (SNPs)
2.2.2 Small insertion or deletion
2.2.3 Copy number variations
2.2.4 Structural variation
2.3 Traditional variant detection methods
2.4 VCF—Variant call format
2.5 Methods of detecting variants
2.6 Limitations of traditions methods
3 Deep learning approach for variant detection
3.1 Convolutional neural network
3.2 Recurrent neural networks (RNNs)
3.3 Deep learning-based variant analysis
4 Variant annotation
5 Current developments and limitations of deep learning in variant detection
6 Conclusion and future scopes
Chapter 19. Inequality in genetic healthcare: Bridging gaps with deep learning innovations in low-income and middle-income countries
1 Introduction
2 Genetic syndrome: Physiological and health implication
3 Comprehensive overview of socioeconomic strains of genetic disorders in LMICs
4 Challenges in genetic syndrome detection and prevention
4.1 LMICs face challenges in accessing genetic screening services and clinical genetic resources
4.2 Limited resources in LMICs lead to delayed diagnosis and preventive care
4.3 Prenatal screening and noninvasive testing are under-implemented in LMICs
5 The role of deep learning in addressing global health challenges
6 Deep learning in genomics: Bridging HIC–LMIC gap in genetic syndrome care
7 Strategies for implementation in LMICs
7.1 Enhancing capacity
7.2 Data accessibility and standardization
7.3 Infrastructural advancements
7.4 Community engagement and collaborative frameworks
7.5 Ethical and regulatory protocols
8 Future directions
9 Conclusion
Declaration of generative AI and AI-assisted technologies in the writing process
Chapter 20. Analysis of genetic and clinical features in neuro disorders using deep learning models
1 Introduction
2 Relationship of WM and GM abnormalities to clinical and genetic features in neuro disorders
3 Related works in machine learning focusing on clinical imaging data for neurodisorder detection
4 Related works in deep learning focusing on clinical imaging data for neuro disorder detection
5 Related works in machine learning focusing on genomic data for neuro disorder detection
6 Related works in deep learning focusing on genomic data for neurodisorder detection
7 Inferences
8 Translational challenges faced when applying findings from model organisms to human neurological disorders
8.1 Biological differences
8.2 Environmental and contextual differences
8.3 Methodological differences
8.4 Ethical and regulatory challenges
9 Conclusion
10 Future works
Index

Life Sciences

Physical Sciences & Engineering

Social Sciences & Humanities

Health

Deep Learning in Genetics and Genomics

Volume 1: Foundations and Introductory Applications

Purchase options

Institutional subscription on ScienceDirect

Khalid Raza