1. Home
  2. NOVA School of Science and Technology
  3. Master in Computational Biology and Bioinformatics
  4. Multi-Omics for Life and Health Sciences
  5. Artificial Intelligence Techniques for Biology

Artificial Intelligence Techniques for Biology

Objectives

This curricular unit aims to provide the student with skills to:

Understand:
- The role of ontologies and symbolic reasoning in biology.
- Supervised and unsupervised machine learning fundamentals.
- Elementary notions of neural networks and deep learning.


Be able to:
- Select and correctly apply the methods and models addressed to biology problems
- Critically evaluate the results obtained.
- Use modern machine learning and deep learning libraries.

Know:
- Supervised and unsupervised learning algorithms frequently used in biology
- Modern AI tools used in biology.

General characterization

Code

12495

Credits

3.0

Responsible teacher

André Francisco Martins Lamúrias, João Alexandre Carvalho Pinheiro Leite

Hours

Weekly - 2

Total - 24

Teaching language

Português

Prerequisites

-

Bibliography

-

Goodfellow, Bengio, Courville, Deep Learning, MIT Press, 2016

Raschka, Liu, Mirjalili, Machine Learning with PyTorch and Scikit-Learn, Packt Publishing Ltd, 2022

Dessimoz, Škunca, The Gene Ontology Handbook, Springer, 2017

Teaching method

The 28 hours of contact between students and teachers will be divided into 14 hours of lectures and 14 of practical tutorial classes.

The theoretical classes will be supported by class notes provided by the lecturer and references to the appropriate chapters of the recommended textbooks. Each of these classes will be divided into approximately 2/3 exposition and 1/3 free discussion with the students.

The tutorial classes will consist both of exercise classes where students follow sets of exercises provided by the tutors as well as project classes where the students can get help for projects of their conception.

Evaluation method

The course unit has two assessment components that contribute to approval and the final grade:

 

Theoretical-Practical Assessment Component – 50% (minimum of 9.5/20)

Project Assessment Component – 50% (minimum of 9.5/20)

The theoretical-practical assessment component consists of a final test or an exam. The project assessment component consists of two mini-projects, initiated during practical classes and carried out in groups. The grade for the project component corresponds to the simple average of the grades of the two projects, rounded to one decimal place.

Attendance to 2/3 of the classes is mandatory for regular students without a justification

Subject matter

1- Introduction to symbolic Artificial Intelligence: knowledge representation by ontologies
2- Semantic Web and applications to biology.
3- Fundamentals of machine learning: supervised and unsupervised; overfitting and model selection.
4- Regression and classification models with artificial neural networks.
5- Introduction to deep learning