Structured Learning Path: ML/DL in Biomedicine
Core training
- Papers
- Top-5 papers in each subfield of interest
- Implementation
- Implement paper and architectures of ML models
- Foundational Mathematics
- Build the mathematical foundation for ML
- Fill the knowledge gaps from reading papers and technical articles
- Foundational Biology
- Molecular Biology
- Genetics
- Proteins
- Advanced ML of field of interest
- Multimodal ML
- Transformer Architecture
- Deep Reinforcement Learning
- Diffusion Models
- ML Systems
- Code
I really need to
- Read papers (to build the foundation, produce the knowledge gaps)
- Read technical articles (to build the foundation, produce the knowledge gaps)
- Python
- Python Basics
- Fluent Python
- Numpy
- Pandas
- Scikit-learn
- Data Engineering
- Feature Engineering
- Data Cleaning
- PySpark
- ML Systems
- Deep Learning Frameworks
- [Book] Mathematics of Machine Learning
- Math Academy
- Linear Algebra
- Calculus
- Probability and Statistics
- Machine Learning Specialization
- Machine Learning Models
- Linear Regression
- Logistic Regression
- Decision Trees
- Model Evaluation
- Accuracy, Loss, ROC AUC, Precision, Recall
- Causal Inference
- Machine Learning Engineering
- Multimodal Machine Learning
- Geometric Machine Learninig
- Deep Learning Specialization
- Deep Learning
- Neural Networks
- Deep Neural Networks
- Convolutional Neural Networks (CNNs)
- VGG
- ResNet
- U-Net
- Recurrent Neural Networks (RNNs)
- Recurrent Neural Network (RNN Lab)
- LSTMs
- GRUs
- Transformers
- [Course] Attention in Transformers
- [Course] How Transformer LLMs Work
- [Notebook] Transformers Encoder from scratch with PyTorch
- [Book] Understanding Deep Learning: Transformers Chapter
- [Book] Transformers and Large Language Models
- [Course] Stanford Language Modeling From Scratch
- [Course] LLM Course by Hugging Face
- Mixture of Experts (MoE)
- Generative AI
- Autoencoders/Variational Autoencoders (VAEs)
- Generative Adversarial Networks (GANs)
- Diffusion Models
- Graph Neural Networks (GNNs)
- Papers
- Deep Reinforcement Learning
- [Course] Deep Reinforcement Learning - Lecture
- Foundations of Deep Reinforcement Learning book
- Genetics Fundamentals
- Molecular Biology
- [Book] Molecular Biology of the Cell
- [Book] Molecular Biology for Computer Scientists
- [Book] Molecular Biology: A Very Short Introduction
- [Book] A Computer Scientist’s Guide to Cell Biology
- [Primer] A Biology Primer for Computer Scientists
- [Graduation] Bioinformática Aplicada a Genômica Médica - Expressão Gênica, Metagenômica e Machine Learning
- [Graduation] Bioinformática Aplicada a Genômica Médica - Análise de Variantes Germinativas e Somáticas
- [Graduation] Medicina de Precisão e Análise de Dados na Saúde
- Cancer Fundamentals
- AI/ML in Healthcare
- [Course] MIT Machine Learning in Computational Biology
- Machine Learning for Healthcare
- AI for Medicine Specialization
- AI in Healthcare Specialization
- Collaborative Data Science for Healthcare
- Data Analytics and Visualization in Health Care
- [Course] Deep Learning for Healthcare Specialization: labs
- [Graduation] Medicina de Precisão e Análise de Dados na Saúde
- Introduction to Applied Biostatistics: Statistics for Medical Research
- Genomic Data Science
- Bioinformatics Data Preprocessing
- NGS data formats (FASTA/FASTQ/VCF)
- Quality control (FastQC)
- Pipelines (alignment, normalization)
- Biopython
- Genomic Data Science Specialization
- [Course] Bioconductor for Genomic Data Science
- [Course] Statistics for Genomic Data Science
- [Book] Bioinformatics Algorithms: An Active Learning Approach
- Introduction to Genomic Data Science
- [Course] HarvardX Biomedical Data Science
- Bioinformatics Data Preprocessing
- AI/ML in Biology for Medicine
- Protein Structure
- Protein Dynamics Simulation
- Scaling deep learning for materials discovery
- DiffDock: Diffusion Steps, Twists, and Turns for Molecular Docking
- BioReason-Pro: Advancing Protein Function Prediction with Multimodal Biological Reasoning
- Accurate structure prediction of biomolecular interactions with AlphaFold 3
- BioEmu: AI‐Powered Revolution in Scalable Protein Dynamics Simulation
- Experimental Data Driven AI Framework for Flexible Protein Conformational Reconstruction
- Free Energy Calculation Method Based on Enhanced Sampling of Diverse Protein Conformations Predicted by Artificial Intelligence
- MELD in Action: Harnessing Data to accelerate Molecular Dynamics