Deep Learning for Evolutionary Genomics (DLEG01)

Course Description

Deep learning has emerged as one of the most powerful approaches for extracting biological insights from large and complex datasets. In ecology and evolutionary biology, deep learning methods are increasingly being applied to genomics, metagenomics, microbiome research, functional annotation, and ancient DNA analysis. This course provides a practical introduction to modern deep learning techniques and their applications to biological data.

Participants will learn the theoretical foundations of neural networks and gain hands-on experience applying deep learning methods to real-world problems in ecology and evolutionary biology. Topics include feed-forward neural networks, convolutional neural networks (CNNs), recurrent neural networks (RNNs), long short-term memory (LSTM) networks, autoencoders, and transformer architectures. Through examples drawn from microbial ecology, population genomics, metagenomics, functional genomics, and ancient DNA research, participants will learn how deep learning can be used to identify biological patterns, classify genomic sequences, infer population history, and generate biologically meaningful sequence data.

• Closest companions to DLEG01
• Machine Learning for Evolutionary Genomics (MLEG01)
• Machine Learning for Evolutionary Genomics: learn predictive modelling, data analysis, and AI methods for genomic data.
• This is the natural prerequisite and companion course. While DLEG01 focuses on neural networks and deep learning architectures, MLEG01 provides the broader machine learning foundations that underpin many deep learning workflows. If you were taking both, I’d generally recommend MLEG01 first, followed by DLEG01.
• Population Genomics and Structure Analysis (PGSA01)
• Learn population structure analysis for evolutionary biology using genetic data, population genetics methods, and evolutionary inference.
• Deep learning methods are increasingly being applied to population structure inference, admixture analysis, demographic history, and evolutionary prediction. PGSA01 provides much of the biological context that makes deep learning applications in genomics meaningful.
• Genome-Wide Association Studies (GWAS) for Evolutionary Biology (GWAS01)
• GWAS for Evolutionary Biology: learn genome-wide association studies, quantitative trait analysis & genomic methods for evolutionary data.
• Many deep learning applications in genomics revolve around genotype–phenotype prediction, feature extraction, and identifying complex relationships within large genomic datasets. GWAS provides one of the most important application domains for these methods.
• Single cell RNA-Seq analysis (SCRN03)
• Learn single cell RNA-Seq analysis with Seurat, 10x Genomics, and advanced QC methods. Gain cell type-specific insights in this live online course.
• Single-cell datasets are among the most common biological datasets currently analysed using deep learning approaches. The course provides exposure to high-dimensional biological data, clustering, dimensionality reduction, and modern computational genomics workflows.

• Strong supporting courses
• Advanced Python for Bioinformatics (APYB02)
• Take your Python skills further. Learn OOP, testing, and optimisation for complex bioinformatics tasks.
• Deep learning research is overwhelmingly Python-based. If your background is primarily in R, this is one of the strongest supporting courses you could take. It helps bridge the gap between biological research and modern AI workflows.
• Introduction to Snakemake (SNKM03)
• Learn Snakemake to automate data workflows. Build reproducible, scalable pipelines for research with hands-on training in this 4-day live online course.
• Deep learning projects often involve complex pipelines containing preprocessing, model training, evaluation, simulation, and post-processing. Learning workflow management becomes increasingly valuable as projects grow in size.
• RNA-Seq Analysis (RNAA02)
• RNA-Seq analysis training – live online course covering experiment design, data QC, alignment, gene expression, DESeq2 differential expression, PCA, visualisation, and functional analysis.
• If you’re interested in applying machine learning to gene expression datasets, differential expression, or transcriptomics, this is one of the most relevant domain-specific courses. It covers DESeq2 workflows, PCA, visualisation, and functional analysis.
• Bayesian Modelling Using R-INLA Course (BMIN04)
• Learn Bayesian modelling with the R-INLA package. Build, fit, and interpret INLA models, define priors and latent effects, and apply INLA to real data in a five day course.
• Not genomics-specific, but highly relevant if you want a broader statistical modelling toolkit. Many researchers working in evolutionary genomics eventually combine ML methods with Bayesian inference and hierarchical models.
• Spatial Phylogenetics and the Bayesian Phylogenetic Mixed Model (PMM) (BPMM01)
• Learn spatial phylogenetics and Bayesian phylogenetic mixed models in R. Practical online training in phylogenetic modelling and comparative workflows.
• This is particularly relevant if your interests lie at the interface of genomics, phylogenetics, comparative biology, and evolutionary inference.

• Recorded courses
• Machine Learning using Python (MLUPPR)
• Machine learning using Python: supervised and unsupervised learning, neural networks, and practical model building with scikit-learn and TensorFlow.
• Deep Learning Using Python (DLUPPR)
• Deep learning course using Python and PyTorch. Learn neural networks, CNNs and transformers through hands-on coding and real data.
• Python for Data Science and Statistical Computing (PYDSPR)
• Learn Python for data science and statistical computing. Build skills in NumPy, Pandas and visualisation across two days of hands-on training for researchers and analysts.
• Deep Learning using R (DLURPR)
• Learn deep learning in R using the torch ecosystem. Build MLPs, CNNs and transformer models through hands-on coding and gain practical skills for real research workflows.
• RNA-Seq Analysis (RNAAPR)
• RNA-Seq analysis training – recorded course covering experiment design, data QC, alignment, gene expression, DESeq2 differential expression, PCA, visualisation, and functional analysis.

What You’ll Learn