András Aszódi
Peter Sarkozy
Contact hours
2x2 hours/week
4 credits

Short Description of the Course:
The Computational Biology and Medicine (CBM) program of AIT helps create a new breed of computer experts who can apply computational and analytical methods to solve complex problems in biomedical research. The CBM course offers a study program that introduces the students to computational biology, with an emphasis on major high-throughput -omics methodologies and databases. The main focus is on the development and application of mathematical modeling and computational simulation techniques for studying biological systems in health and disease.

In the first half of the course gene expression regulation, qualitative and quantitative description of complex kinetic phenomena, metabolic control, information processing in biochemical reactions, homeostasis and robustness will be discussed in detail. In the second half we focus on the data analysis tasks associated with  high-throughput genomic sequencing experiments, with special emphasis on medical decision support.

Motivation and perspective: 
The application of systems biology and computational medicine to translational research in the pharmaceutical and biotech industries is one of the most important recent developments in computational biology. Building on applications of computer science in the field of biology, bioinformatics research requires input from the diverse disciplines of mathematics and statistics, physics and chemistry, and medicine and pharmacology. AIT students interested in computational biology and medicine will be introduced to this multidisciplinary perspective and its applications in academic and industrial environments.


  • Calculus, in particular differential equations, Fourier transform methods.
  • Linear algebra: vectors, matrices, eigenproblems.
  • Basic probability theory: discrete distributions, Markov chains, Bayes’ Rule and its applications.
  • Python programming skills. Knowledge of SageMath is an advantage.
  • High-school biology.


Topics in detail

Part I: Biocybernetics and Systems Biology (A. Aszódi)

Introduction to biocybernetics
Definition of systems. Comparison of natural and artificial systems. Applicability of systems theory and engineering in biology. Components of biological systems. Biological time series. Modelling biological phenomena.

First-order kinetics
Stock-and-flow model of inflation. First-order chemical reactions and reaction networks. Systems of linear ordinary differential equations. Solving linear ODE systems by finding the eigenvalues and eigenvectors of the coefficient matrix. Qualitative behaviour of linear systems.

Biochemical kinetics
Bulk-phase mass action kinetics. Reversible reactions, equilibrium. Isomerisation catalysis. Enzyme kinetics models: steady-state approximation, Michaelis-Menten rate equation. Competitive and noncompetitive inhibition.

Principles of genomic regulation
The molecules of life: DNA, RNAs, proteins. The "central dogma": transcription and translation. The differences between prokaryotic and eukaryotic organisms. Prokaryotic genome regulation: the E. coli lac operon, the lambda phage bistable switch, the "Repressilator" synthetic genetic network. Eukaryotic genome organisation: chromosome structure, splicing, epigenetic regulation. The self-regulatory Hes1 network.

Oscillations and chaos
Periodic phenomena in living systems. The Byelousov-Zhabotinsky chemical oscillator. Biological examples of oscillatory phenomena: glycolysis, mitotic oscillations, predator-prey interactions (Lotka-Volterra models). Chaotic dynamics. The Lorenz attractor and the logistic map. Chaos in enzyme-catalysed reactions. Qualitative behaviour of nonlinear differential equations, the linearisation approach. Detecting chaotic behaviour with Lyapunov exponents.

Stochastic biochemical kinetics
Basic probability theory. Rolling dice in software. Bayes' Rule. Principles of stochastic kinetics. Master equation approach, the Gillespie stochastic simulation algorithm. Properties of Markov chains. "Convergence" of stochastic kinetic trajectories to the bulk kinetics model in the limit of very large number of molecules.

Systems modelling
How to construct mechanistic hypotheses from observations. Popper's falsifiability theory. Kinetic indistinguishability. System analysis by structural or parameter perturbation. Robustness of genetic networks. Self-regulation in biochemical networks: metabolic control theory, flux balance analysis.

The theory of evolution
Fundamental concepts. Lamarckian and Darwinian evolution. Evolution of macromolecular sequences, molecular phylogeny. Epigenetic inheritance.

Computing with biomolecules
Quasi-digital approaches: Adleman's DNA-based solution of the travelling salesman problem and related efforts. Molecular implementations of Boolean logic gates. Computing with enzymatic reaction networks. Simple learning phenomena.

Regulation in spacetime
Algorithmic models of plant growth, applications in computer graphics. Turing's theory of morphogenesis. Modelling reaction-diffusion networks with partial differential equations. Solving the diffusion equation with Fourier transformation techniques. The Gierer-Meinhardt model of pattern formation in Dictyostelium discoideum. Robustness of pattern formation in living systems.

Artificial life
Chemoton theory: self-reproducing autocatalytic reaction networks. Cellular automata, Conway's "The Game of Life". In silico models of simulated evolution: the Tierra and Avida systems.


Part II: Data analysis of high-throughput genome sequencing (P. Sarkozy)

Introduction to molecular genetics
Role and characteristics of DNA in organisms. Mutation types, population genetics, linkage disequilibrium, transcription and translation of DNA to proteins, gene expression, epigenetic modifications, the Human Genome Project and the path to personalized medicine.

Overview of DNA sequencing technologies
Sanger sequencing to single-molecule real-time DNA sequencing, in vitro diagnostics, high-throughput measurement methods, partial genetic association studies, genome-wide association studies, single-molecule real-time DNA sequencing.

High-throughput measurements
Quality control, filtering, common failure modes and platform-specific error profiles of common measurement methods, sample multiplexing and study design.

Strings in Bioinformatics
Naïve exact matching, Z algorithm, naïve approximate matching, radix sorting, suffix indices, longest common prefix, Burrows-Wheeler transformation.

Mapping and assembly of large, complex genomes
Alignment scoring schemes, de-novo assembly, reference mapping, the Smith-Waterman algorithm, the Needleman-Wunsch algorithm, understanding and correcting alignment bias in DNA sequencing, local and global alignment, paired-end sequencing.

Phylogenetics and metagenomics
Multiple sequence alignment, clustering approaches, distance metric, phylogenetic tree construction, metagenomic population studies, molecular clock hypothesis.

Interpretation of results
Identifying variants, detecting somatic mutations, heterogeneous population sequencing, construction of local phylogenetic trees for cancer evolution, resolving haplotypes, copy number variations, large-scale genomic rearrangements.

DNA editing
DNA repair mechanisms, homology directed repair, DNA editing in-vivo and in-vitro, CRISPR-CAS9 system, zinc finger nuclease technology, in vivo delivery methods, RNA interference, system biological approach to diseases.

Models in Computational Genomics
Markov chains, Gaussian mixture models, hidden Markov models, support vector machines, biologically inspired artificial neural networks, neural network training, convolutional neural networks, feedforward-backpropagation algorithm.

Bayesian frameworks in Bioinformatics
Frequentist vs. Bayesian approaches, naïve Bayes classifiers, Bayesian networks, probabilistic classifiers, network structure learning, Semantic technologies for computational biology, drug

The co-evolution of man and machine
Brain-computer interfaces, the challenges of biosynthetic organisms, the extension of mankind with weak artificial intelligence, challenges proposed by strong artificial intelligence.



Part I:
Data analysis and simulation tasks (reaction kinetics of biochemical pathways, predator-prey models, biological pattern formation models etc.) using the SageMathCloud on-line computer algebra system.

Part II:
Construction of a phylogenetic Tree of Life from 18S ribosomal subunit sequences, or data analysis of an in-silico genetic association study using various open-source software packages. Simulation of a DNA sequencing experiment, analysis of the results.



Part I:
The students prepare an essay (10 pages minimum) on biological information processing from a topic list provided by the lecturer. Those who propose a topic on their own will get an extra half grade (i.e. B+ instead of B). Instead of writing an essay it is also possible to write small programs to simulate biological regulation phenomena.

Part II:
The students write a research report that summarizes their homeworks and provide an objective overview of their results in no less than 10000 characters in length (without spaces). An additional grade (e.g.  B to A, B+ to A+ ) will be given for the use of publicly available real measurement data instead of simulated data.

Grading Criteria:

  • Essay and research report (70%): the students must demonstrate that they have understood the principles discussed in the lectures and can apply their knowledge in a practical context. Originality and a critical approach is especially important.
  • Course activity (20%): students are required to ask questions and challenge the lecturer and each other.
  • Homeworks (10%): timely completion of the tasks with correct results is required.


Instructors' bio:

András Aszódi (born 1964) studied chemistry at Eötvös Loránd University in Budapest where he graduated in 1988. He then studied molecular neurobiology at the University of Oxford, supported by a Soros scholarship. He received his Ph.D. in 1991 on the kinetic models of simple learning processes. From 1992 to 1996 he developed protein structure prediction methods at the National Institute for Medical Research in London. In 1996 he joined the Novartis Research Institute in Vienna as a computational modeller. He built up the In Silico Sciences unit that provided bioinformatics and computational chemistry tools to researchers. In 2006 he joined the Research Institute of Molecular Pathology in Vienna where he was developing data analysis tools and databases for high-throughput sequencing projects. He is currently working in the BioComp group of the CSF GmbH and also teaches a systems biology course at the University of Vienna. He has over 35 scientific publications, including a book with W.R. Taylor on protein structure prediction.

Peter Sarkozy (born 1984) received his degree in Computer Science from the Budapest University of Technology and Economics in 2009, and continued his graduate studies at the Department of Measurement and Information systems. During his graduate studies from 2009 to 2012 he participated in multiple projects together with the Department of Genetics, Cell and Immunobiology at the Semmelweis University. His areas of interest include the measurement and error characteristics of next-generation DNA sequencing technologies. He is the first person in Hungary to apply Oxford Nanopore Technologies’ single molecule real-time sequencing technology. He is currently working as a research assistant at the Department of Measurement and Information Systems at BUTE.