General provisions

The course is given as a stand-alone course. The course is Internet-based and requires access to a computer with Internet connection.

Entry requirements

150 credits in chemistry, biology, biochemistry, pharmacy, medicine or dentistry and at least 4 credits from Pharmaceutical Bioinformatics

Learning outcomes

After completing the course, the student should:

• Explain predictive chemobioinformatic modeling and show how it can be used to solve problems in drug discovery and life sciences.
• Explain the background to and perform and evaluate modeling with supervised and unsupervised machine learning methods.
• Calculate and use chemical descriptors for proteins, peptides and organic molecules using chemobioinformatic tools.
• Perform cluster analysis and explain how they can be used to solve chemobioinformatic problems in the pharmaceutical field.
• Independently be able to build simpler QSAR and proteochemical models, validate and interpret them, and be able to apply the models to solve problems in the life sciences and the pharmaceutical field.

Content

The course provides an overview of statistical modeling methods with applications in the field of medicine, life sciences and pharmacology and focuses on practical exercises where problems are solved by the different methods.

The course specifically includes:

• Introduction to statistical modeling in pharmaceutical bioinformatics.
• In-depth review of QSAR and proteochemometry as well as descriptors for proteins, peptides and organic molecules.
• Review of supervised and unsupervised methods for statistical modeling / analysis such as PCA, PLS, SVM, random forest, kNN.
• Review of cluster analysis and experimental design and methods / tools for this.
• Introduction to the statistical programming language R.
• Practical exercises in calculating descriptors for protein and peptide sequences as well as organic molecules.
• Practical exercises in cluster analysis.
• Practical exercises in the use of PCA, PLS, SVM, random forest and kNN.
• Practical exercises in building QSAR and proteochemometrical models.

Instruction

The course is given as a distance course using the Internet and access to a computer with Internet connection is compulsory. The work is done individually through self-study of web-based materials and computer exercises. The reading material is interspersed with interactive questions. Communication between student and teacher takes place via web and email where the student has the opportunity to ask questions. Mandatory elements are interactive questions, which are answered via the website. The course is given in English at half-time (50%).

Assessment

Written exams take place at the end of the course. For an approved course, in addition to a passing exam (4 hp), a pass result is required on compulsory parts (1 hp). Opportunities to supplement unapproved compulsory parts is offered earliest at the next course opportunity. If there are special reasons, the examiner may make exceptions to the specified examination method and allow a student to be examined in another way. Specific reasons may be e.g. need of special educational support verified by the university coordinator.

Other directives

The course replaces and corresponds to 3FF777 Applied Pharmaceutical Bioinformatics, 5.0 higher education credits.