Advanced Methods in Bioinformatics

6 ECTS credits
180 u studietijd

Aanbieding 1 met studiegidsnummer 4018221FNR voor alle studenten in het 2e semester met een gespecialiseerd master niveau.

Semester

2e semester

Inschrijving onder examencontract

Niet mogelijk

Beoordelingsvoet

Beoordeling (0 tot 20)

2e zittijd mogelijk

Onderwijstaal

Engels

Onder samenwerkingsakkoord

Onder uitwisselingsakkoord mbt studiedelen

Faculteit

Faculteit Wetenschappen en Bio-ingenieurswetensch.

Verantwoordelijke vakgroep

Computerwetenschappen

Onderwijsteam

Wim Vranken (titularis)

Onderdelen en contacturen

26 contacturen Hoorcollege
26 contacturen Werkvormen en Praktische Oef.
12 contacturen Zelfwerk en -studie

Inhoud

This course focuses on algorithms and methods in computational biology, with active participation of the students in implementing and critically analysing published research. Core biological background knowledge is provided through targeted lectures and an AI-supported biological background question agent, designed to answer domain-specific questions from a computational perspective.
Students reimplement a real scientific article, with particular attention to understanding the underlying datasets (e.g. data splits, bias, structure) and exploring computational methodology (e.g. reparameterisation, alternative modelling choices). Interactive in-class toy problems are used to expose common pitfalls in biological data analysis before formal methods are introduced.

The course covers methods and associated algorithms for solving problems related to:

i) protein sequence, structure and evolution, including alignment and protein language models

ii) Variant effect prediction (protein and DNA)

iii) Single-cell RNA-seq and cancer genomics

The aims of the course are to:

i) familiarise students with the background and concepts in these two fields

ii) give students insight in how data, computational methodology and results are interconnected in Computational Biology and Bioinformatics

iii) enable the student to implement algorithms in these fields

iv) critically investigate the results produced by such algorithms. This knowledge should enable the students to understand the literature on bioinformatics and computational biology.

This is achieved through:

a) lectures to transfer the basic concepts, including article fragments to train the student in how to extract information from scientific literature, supplemented by interactive in-class toy problems are used to expose common pitfalls in biological data analysis before formal methods are introduced

b) a project where a relevant scientific article is reimplemented, critically assessed and written up in article form (individual or in student pairs)

c) a peer review stage where students review each others' projects and provide critical feedback and d) a final open exam presentation on the project with questions from peers and instructors.

Studiemateriaal

Handboek (Aanbevolen) : Protein bioinformatics, An Algorithmic Approach to Sequence and Structure Analysis, Ingvar Eidhammer, Inge Jonassen and William R. Taylor, Wiley, 9780470848395, 2004
Handboek (Aanbevolen) : Molecular biology of the cell, Bruce Alberts, Alexander Johnson, Julian Lewis, Martin Raff, Keith Roberts, and Peter Walter, 7e, Garland Science, 9780393884852 , 2023
Handboek (Aanbevolen) : Understanding bioinformatics, Marketa Zvelebil, Jeremy O. Baum, Garland Science, 9780815340249, 2007

Bijkomende info

Students are expected to have intermediate or better skills in programming. All lectures will be provided in hybrid form.

The use of generative AI is explicitly integrated into the course as a learning support tool. Students are allowed to:

use AI tools to provide feedback on self-written text (e.g. structure, clarity, English), but not to generate text from scratch;
use AI assistance for coding, provided the student understands, adapts, and documents the result.

In all cases, students must acknowledge all AI use and remain fully responsible for the content, correctness, and interpretation of their work, in line with VUB guidelines.

Leerresultaten

Algemene competenties

Recognition of key terms in computational biology and bioinformatics, and awareness of algorithms to solve biological or medical questions. Implementation and exploration of an algorithm or method from computational biology or bioinformatics.

Capacity to communicate

Ability to communicate constructively with peers in a joint project, and the clear and succinct presentation of information contained in a scientific article.

Capacity to evaluate algorithm

Understanding the concepts behind computational biology and bioinformatics, and application of this knowledge to algorithms that analyse or predict biological data. This includes awareness of pitfalls in using biological data, such as data quality, and bias or overlap in the data used, including those arising from automated or AI-assisted analyses.

Analyzing results

Analysis of the outcomes of your algorithms, with critical assessment of whether they perform correctly (or not).

Evaluating others

Evaluation of a presentation on a scientific article by your peers, including giving constructive and critical feedback about any concerns you might have.

Synthesize an article

Critical reading and active implementation of a scientific article in this field, with synthesis of the key points and with reflection on how computational tools (including AI) influence interpretation, to be communicated in a presentation.

Beoordelingsinformatie

De beoordeling bestaat uit volgende opdrachtcategorieën:
ZELF Presentatie bepaalt 75% van het eindcijfer

ZELF Paper bepaalt 25% van het eindcijfer

Binnen de categorie ZELF Presentatie dient men volgende opdrachten af te werken:

oral examination met een wegingsfactor 75 en aldus 75% van het totale eindcijfer.

Binnen de categorie ZELF Paper dient men volgende opdrachten af te werken:

project report as manuscript met een wegingsfactor 25 en aldus 25% van het totale eindcijfer.

Aanvullende info mbt evaluatie

The evaluation is based on a project where an existing published method from the field is re-implemented and presented (at the exam, individual or in groups of two). Grades are assigned through the report (25%) and the exam presentation (75%). The evaluation explicitly rewards:

- the quality of the project in terms of the implementation and the analysis of the results, including documented and justified methodological choices;

- the presentation of the article;

- critical insights by the student, including critical reflection on data quality, assumptions, and model limitations;

- quality of the peer review and participation during the exam;

- transparent and responsible use of AI tools.

AI-generated content without understanding or attribution is penalised, and all source material has to be correctly referenced: plagiarism of code or text is not allowed and will result in a zero score and possible disciplinary sanctions.

Toegestane onvoldoende

Kijk in het aanvullend OER van je faculteit na of een toegestane onvoldoende mogelijk is voor dit opleidingsonderdeel.

Academische context