SGN-53007 Computational Diagnostics, 5 cr

Additional information

Suitable for postgraduate studies.

Person responsible

Frank Emmert-Streib

Lessons

Implementation Period Person responsible Requirements
SGN-53007 2016-01 1 Frank Emmert-Streib
To complete the course, the student is required to (all three requirements must be completed to pass the course): a) Execute the project work (20% of the final grade) b) Execute the weekly exercises (1 per exercises lesson, 40% of the final grade) c) Do the final exam (40% of the final grade)

Learning Outcomes

After completing the course, the student gained a basic understanding of the definition and the meaning of computational diagnostics and its utility for biomedical research and business intelligence. Case studies will be discussed illustrating the interplay between computational and statistical methods that are applied to large-scale and high-dimensional data sets from genomic experiments and business processes, e.g., the stock market or consumer behavior. Moreover, the student will learn how to practically approach such problems by using the statistical programming language R. In general, the course teaches statistical thinking in the context of biomedical and business problems, i.e., the adaptation of machine learning methods in a problem specific manner.

Content

Content Core content Complementary knowledge Specialist knowledge
1. Classification of disease groups  Computational implementation and interpretation; classification methods   
2. Business intelligence  Data analysis methods for business processes   
3. Survival analysis  Regression models for time-to event processes   
4. Genomics data  Preprocessing and normalization of gene expression data from microarray experiments   
5. Programming in R  Usage and programming in the statistical programming language R   
6. Quantitative assessment of results  Statistical error measures; resampling techniques   
7. Predictive models  Linear regression, hypothesis testing; general models in data science   

Instructions for students on how to achieve the learning outcomes

To complete the course, the student is required to (all three requirements must be completed to pass the course): a) Execute the project work (20% of the final grade) b) Execute the weekly exercises (1 per exercises lesson, 40% of the final grade) c) Do the final exam (40% of the final grade)

Assessment scale:

Numerical evaluation scale (0-5)

Partial passing:

Completion parts must belong to the same implementation

Study material

Type Name Author ISBN URL Additional information Examination material
Book   An Introduction to Statistical Learning   Gareth James, Daniela Witten, Trevor Hastie and Robert Tibshirani       Introductory overview of many methods discussed in the lectures.   No   
Book   Statistics and Data Analysis for Microarrays Using R and Bioconductor   Sorin Drăghici       Introduction to the analysis of microarray data.   No   

Additional information about prerequisites
Basic programming skills. Experience with the language R are desirable, but not necessary. Basic knowledge in Mathematics and Machine Learning. Basic knowledge of biology/systems biology.



Correspondence of content

Course Corresponds course  Description 
SGN-53007 Computational Diagnostics, 5 cr BMT-53007 Computational Diagnostics, 5 cr  
SGN-53007 Computational Diagnostics, 5 cr SGN-53006 Computational Modeling in Biomedical Problems, 5 cr  

Updated by: Värri Alpo, 08.04.2016