Data Science and Biology

This course focuses on applying data science methods to biological data. With the rise of high-throughput techniques in biology, massive datasets are now common, including genomes, metagenomes, transcriptomes, and more. The course equips students with the theoretical and practical skills to extract insights from such data. Students learn to work with the command-line interface, write scripts in Python and bash, and use data analysis and visualization techniques. They also explore supervised and unsupervised machine learning methods and design computational pipelines to answer biological questions using large datasets. The course combines online modules, tutorials, group projects, guest lectures, and practical assignments.

Program/Minor

Learning Outcomes

Understand file systems, data types, and command-line usage; Write scripts in Python and bash to analyze biological datasets; Select and justify appropriate data analysis and visualization methods; Apply and evaluate basic machine learning approaches; Design publication-ready scientific figures; Develop computational pipelines to answer biological questions; Document analyses for full reproducibility; Collaborate effectively in a research team; Critically analyze and report data science results; Conduct a complete research cycle from question to written report.

Prerequisites

Completion of the course "Genomica" or equivalent knowledge; Basic knowledge of statistics (e.g. from "Biological models and statistics"); Not recommended for MBLS students due to content overlap with “Bioinformatics and Dynamical Modeling (MBLS-207)”

Details

Institution: UU

Period: Period 3

Level: Intermediate

EC: 7.5 ec

Data Science Ethics Colloquium Series

Ethical and legal considerations are essential to many applications of data science.
This colloquium series is intended to make the students aware of such considerations and to give them the vocabulary to discuss those matters with experts of legal and ethical aspects.
Furthermore, the colloquium will enable students to develop an understanding of professional integrity and provides them with applicable practices for conducting responsible data science.

Students develop an understanding for the ethical concerns of data science. They recognize how data and data practices, machine learning and their implementation into business models or public administration is related to societal values. Students identify various ethical challenges raised through current data practices and their application in society.
Students will identify different layers of responsibility for designing data science applications, applying them in societal contexts and regulating them. They are familiar with case examples of best and worse practices, and texts addressing ethical, legal, political and social concerns of data science.

Program/Minor

Learning Outcomes

After this colloquium, students will be able to identify ethical issues in data projects; They can recognize responsibilities and voice their own concerns and how these affect design decisions; They are familiar with guidelines for responsible data ethics and the code of conducts of their profession; The objective of this colloquium is to help students to develop a professional integrity as responsible data scientists.

Details

Institution: UU

Period: Period 1

Level: Master

EC: 1 ec

Data Science for Plant Breeding and Genetics

One of the major goals for plant breeders is to identify candidate varieties that are well adapted to the set of environmental conditions that are relevant for the agricultural system of interest. To achieve this goal, breeders characterize their genotypes in multi-environment trials or in phenotyping platforms. The ranking of genotypes might change across trials, indicating that different genotypes will be well adapted to different environmental conditions. This phenomenon is called genotype-by-environment interaction (G×E) and provides an opportunity to increase the sustainability of agricultural systems by using varieties that will perform well in specific growing conditions. For example, breeders might want to develop varieties that are well-adapted to drought or to specific agronomic management conditions (e.g. organic agriculture).
Thanks to the new genotyping and phenotyping technologies, breeders are facing the exciting challenge of how to integrate multiple data sources, generating predictions that will inform selection decisions. Such predictions can be made for yield performance, in combination with other parameters like adaptability, plasticity sensitivity and stability. The main goal of this course is to provide students with data science and statistical tools to analyse single – and multi-environment trials, generate predictions for the genotypic response to the environment and visualize the genotypic response across environments.
For the analysis of single-environment trials, this course will discuss about the principles of experimental design and illustrate them by means of commonly used designs as RCBD, lattice designs, alpha designs and row-column designs. Issues related to the estimation of heritability will be discussed. BLUPs and adjusted means will also be addressed. Phenotypes across multiple environments will be modelled focusing on G×E, using models with different levels of integration of biological knowledge. The course will start with a simple variance components models, and continuing with linear-bilinear models and factorial regression models. Later, G×E will be discussed from the perspective of QTL and genomic prediction models. Finally, the use of models with a larger level of biological integration (e.g. crop growth models) will also be discussed. The utility of these models to classify genotypes and environments will also be discussed.

Program/Minor

Learning Outcomes

Describe in general terms the commonly-used and novel systems to collect phenotypic data (multi-environment trials, platforms, imaging) and how do these systems play a role in characterizing plant adaptation; Describe the basic principles of experimental design and apply them (CRD, RCBD, incomplete block designs, alpha and row-column designs, p-reps, split plot designs); Explain the general principles of ANOVA and mixed models, fit a variance components model by REML (using R) and interpret the output to characterize G×E in terms of .genotype by location, genotype by year, genotype by management, etc.; Formulate linear-bilinear models (Finlay-Wilkinson, AMMI and GGE), use R to fit the models, and interpret the general output from the analysis, the estimated parameters and relevant (bi-)plots; Explain the principles of single- and multi-environment genomic predictions and formulate the corresponding mixed models;- explain G×E in relation to consistency of QTLs; Discuss the utility of crop growth models to characterize environments over time

Prerequisites

Advanced Statistics or Data Analysis for Biosystems Engineering or Advanced Statistics for Nutritionists

Details

Institution: WUR

Period: Period 6

Level: Advanced

EC: 6 ec

Datagedreven werken bij maatschappelijke opgaven, introductie tot AI geletterdheid, Systeemdenken in Landbouw en Voedselsystemen

Details

Institution: HAS Green Academy

EC: LLO ec

Deep Learning 1

Deep learning is primarily a study of multi-layered neural networks, spanning over a great range of model architectures. We cover the basic following content in theory and in practice. We adapt the content, especially in the last lectures, based on the latest advances:

Introduction to deep learning. A brief introduction of deep learning. History, recent trends.
Deep feedforward network. Neural networks are complex cascades of simple building blocks. We learn the basic idea that creates complexity out of simplicity.
Optimisation. Since neural networks are notoriously difficult to train, we will deepen our understanding of how to optimise them both theoretically and practically.
Deep Learning Regularisation. A central problem in machine learning is how to make an algorithm that will perform well not just on the training data, but also on new inputs. Many strategies used in machine learning are explicitly designed to reduce the test error, possibly at the expense of increased training error. These strategies are known collectively as regularisation.
Convolutional neural networks (CNN). The driving force behind the popularity of deep learning. CNNs have their main application in image recognition, object detection, automatic text translation, and speech recognition. We learn the basics as well as the latest models that are used by all academia and the tech giants.
Modern convolutional neural networks. These models include AlexNet, the first large-scale, modern convolutional network that wins the champion on a large-scale vision challenge; the VGG network (VGGnet), which is even deeper and makes use of a number of repeating blocks of elements; the GoogLeNet, which makes use of networks with parallel concatenations (GoogLeNet); residual networks (ResNet) which are currently the most popular go-to architecture today, and densely connected networks (DenseNet), which are expensive to compute but have set some recent benchmarks.
Transformers. The transformer model is a seq2seq model which uses attention in the encoder as well as the decoder, thus eliminating the need for RNNs, as we explain below. Transformers have been used for many (conditional) sequence generation tasks, such as machine translation, constituency parsing, music generation, protein sequence generation, abstractive text summarisation, image generation (treating the image as a rasterized 1d sequence), etc.
Graph neural networks. Traditionally, neural networks are applied to data with a well-specified and orderly structure, like images or text. With graph neural networks we learn how to harvest the power of deep learning on graph data, such as graphs extracted from social media, molecular graphs, and beyond.
Generative learning. One of the most impressive outputs of Deep Learning is algorithms that paint very realistic but fake data, such images of faces that never existed. Here, we will learn about different generations of generative learning algorithms, along with their connections and open problems.
Self-supervised Learning. Learning from raw data without annotations provides benefits as these methods can be applies to extremely large-scale data and offer other benefits. Starting from NLP, the methods in computer vision have tremendously matured and are diffusing into multiple applications.
Multi-modal Learning. Learning from multiple modalities has shown extremely promising results, especially for naturally paired data such as audio-video and human-constructed paired data such as images and their captions. We will learn about current approaches in this domain.
Open questions in deep learning. Deep learning continues to advance, but many questions remain open. We will discuss open questions ranging from continual learning to adversarial attack, as well as the open research questions regarding the geometric foundations of deep networks.

Program/Minor

Learning Outcomes

The students can explain and motivate the fundamental principles and mechanisms behind Deep Learning’s past, present and future; The students can explain the major challenges, directions and active domains of research in the field of deep learning along with their advantages and disadvantages; The student can program, train and run deep learning models in a server environment; The student can express research ideas described in publications in clear terms for readers unfamiliar with the details and devise promising follow-up research directions; The student is able to debug and critically assess deep learning methods from a practical-engineering, and mathematical-theoretic point of view; The student can independently tackle new deep learning problems with well-reasoned combinations of existing approaches.

Prerequisites

Machine Learning 1 (this does not apply to first year AI students) Recommended prior knowledge: Linear Algebra, Calculus, Basic Probability Theory, Basic Machine Learning, Programming, especially Python, Machine Learning 1

Details

Institution: UvA

Period: Period 2

Level: Master

EC: 6 ec

Design thinking, leiderschap in duurzaamheid, transitions/interdisciplinary collaboration

Life-long learning: Courses of 4-10 weeks with in-person meetings

Details

Institution: UU

EC: LLO ec

Digital innovation

In the course Digital Innovation we focus on the role of Information and Communication Technology (ICT) in the fundamental transformations of industries and societies. After introducing the nature of ICTs, the first part of the course starts by providing an historical and theoretical introduction to digital innovation from a societal perspective. Specific emphasis is on the economic and social effects of digitalisation. The course then goes into recent developments brought about by digital platforms and artificial intelligence. Throughout the course, attention is paid to the role of digital innovation in specific domains such as health and mobility. In the second part of the course students apply their knowledge to a case in a domain of their choice.

Program/Minor

Learning Outcomes

Understand the basic technological principles underlying ICTs, recent developments in ICTs, and their impacts on economy and society; Understand innovation processes that lead to new digital products and services, both in the private and the public sector; Understand the effects of robotization on work and the labour markets; Understand the role of digital platforms and the sharing economy in redefining marketplaces; Evaluate the possibilities and impacts of Artificial Intelligence; Reflect critically on the opportunities and threats of digital technology, and on rationales and government policies in the digital domain; Reflect critically on the impact of digital innovations on the individual, organizations, and society at large; Explain and apply theoretical frameworks relevant to the analysis of digital innovations

Prerequisites

Assumed: Basic knowledge of innovation theory (can be gained by following the course Introduction into Technology and Innovation (GEO1-2201))

Details

Institution: UU

Period: Period 4

Level: Advanced

EC: 7.5 ec

Digitale transformatie met AI, toegepaste data science, AI geletterdheid

Life-long learning: Both larger training programs and short course days. You can enter this course with your own AI questions

Details

Institution: Inholland

EC: LLO ec

Ecogenomics

The field of ecological genomics strives to uncover the genetic and molecular mechanisms influencing responses and adaptations of organisms to their environment. Achieving this aim requires insight in evolution and selection pressure and how that results in natural variation. Using this natural variation to study mechanisms requires a good understanding of both ecologically important phenotypes and species with a well-developed genomic tool-kit. Molecular biological and genomic tools have primarily been developed for model organisms (such as yeast and Arabidopsis) representing a narrow spectrum of phenotypes, whereas many organisms that are the focus of ecological research have had limited genomic resources devoted to them.

In this course, we will focus on the ways in which ecogenomics unites genomic and ecological approaches, which are: (1) to explore the natural variation found within model or crop species, (2) to analyze close relatives with a broader range of phenotypes (e.g. comparative genomics ), and (3) to generate new genomic resources for the species of interest . We will explore the underlying principles of genomics and how they are being applied for each of these three approaches. We will specifically focus on ecogenomics studies of abiotic stress responses. Specific approaches that will be discussed and demonstrated include: genetic mapping, expression analysis (transcriptomics), phenotyping (metabolomics, microbiome analysis, etc.), gene candidate verification. Analysis tools that will be studied “hands-on” will include: the use of genomic databases and tools, quantitative trait loci (QTL) analysis, physiological experiments and genome-wide association studies.

Program/Minor

Learning Outcomes

Explain the strategy of Ecogenomics in answering biological questions; Explain the principles of and difference between association mapping (GWAS) and mapping in biparental populations (QTL mapping); Explain the principle of genomics, transcriptomics, metabolomics, microbiome analysis; Discuss the principles of how organisms can adapt or respond to their environment; Translate knowledge on genetics and omics into ecological insight or breeding strategies; Use online genomics resources; Carry out an ecogenomics research project; Analyse the results of a Genome Wide Association Study; Interpret and present the experimental results

Prerequisites

The course is meant for 3rd year (major) Biology students who have successfully completed the 2nd year courses Cellulaire Responsen (5042CERE6Y) and Challenges in Molecular Life Science (5042CIML6Y) or equivalent. Content of the other 2nd year UvA Biology courses will also be considered as known. Students from outside the UvA, or those who followed another BSc program, should contact the course coordinator before registration.

Details

Institution: UvA

Period: Period 2

Level: Intermediate

EC: 6 ec

Ethics and Philosophy for Biologists

This course deals with biology as a science and with moral issues in society that are related to biology. Questions on these topics often don’t have definite answers. The aim of this course is to explore such open questions in two domains of philosophy. We will deal with:
1. Philosophy of science. To understand biology as a science, we discuss the philosophical question of what science is. What does it mean to do scientific research today? What makes knowledge ‘scientific’? Interviews with researchers in biology will add concrete and topical input for thinking.
2. Ethics. Many developments in biology raise ethical questions. How should we treat nature and non-human animals? Do we have a moral duty to safeguard biodiversity? How to use genetic modification tools in a morally responsible way? What to think of biological explanations of moral behavior? Morality is both all-pervasive and elusive. Considering a diversity of ethical approaches helps to get a grip on these issues.

Program/Minor

Learning Outcomes

Analyze (moral) problems in the context of contemporary scientific research; Describe central concepts in the philosophy of science in relation to biology; Understand that the place and function of biology is to be understood in its societal context; Recognize logical fallacies and the structure of argumentation in texts; Apply ethical theories and concepts to a variety of moral issues related to biology; Critically reflect on their own values and explore how these may be adaptable to change

Prerequisites

n/a

Details

Institution: WUR

Period: Period 5

Level: Introductory

EC: 3 ec

Ethics in AI design, AI en ML skills

Life-long learning: Multiple free self-paced courses

Details

Institution: TUD LLO

EC: LLO ec

Evolutionary Algorithms

In this course we consider a specific subfield of Artificial Intelligence: Evolutionary Algorithms (EAs). These algorithms, sometimes also identified as being part of the class of bio-inspired algorithms, have as a metaphor the concept of natural evolution, i.e., the mechanisms by which, the fittest individuals in a population survive, reproduce, and in doing so, over time, change to be better equipped to thrive in their environment. Initiated in the 60s and 70s of the 20th century, research on EAs has progressed immensely. Today, EAs are being used to solve real-world problems in many areas, e.g. to optimize the layout of electrical wind farms, to automatically create radiation therapy treatment plans, and to optimize the architectures of deep neural networks.

This course covers a spectrum of topics in EAs, ranging from basic concepts to advanced, recent, and state-of-the-art research, and ranging from theoretical to applied. In particular, topics include genetic algorithms, evolution strategies, genetic programming, estimation-of-distribution algorithms, optimal mixing evolutionary algorithms, multi-objective optimization, (GPU) parallelization, and real-world applications.

Program/Minor

Learning Outcomes

Explain the key concepts underlying the main streams in Evolutionary Algorithm (EA) research, with in particular genetic algorithms, evolution strategies, genetic programming, estimation-of-distribution algorithms, and optimal mixing evolutionary algorithms; Explain key ingredients underlying the rationale of when these algorithms work and when they do not work. In particular: schema analysis and how the match between the search bias of an EA and the fitness landscape is influenced by aspects such as variable dependencies and multi-modality; Explain key research lines along which state-of-the-art research in EAs is done to achieve more robust, efficient, and effective EAs; Identify good opportunities for using EAs, or hybrid versions thereof, in practice; Properly (scientifically) experiment with EAs as well as program your own

Details

Institution: TU Delft

Period: Period 4

Level: Master

EC: 5 ec