MSchim Biostatistics and Statistical Biocomputing

Faculty of Science
Autumn 2005
Extent and Intensity
2/0/0. 2 credit(s) (fasci plus compl plus > 4). Type of Completion: z (credit).
Teacher(s)
Prof. Michael Schimek (lecturer), prof. RNDr. Ivanka Horová, CSc. (deputy)
RNDr. Tomáš Pavlík, Ph.D. (seminar tutor)
Guaranteed by
prof. RNDr. Ivanka Horová, CSc.
Department of Mathematics and Statistics – Departments – Faculty of Science
Timetable of Seminar Groups
MSchim/01: No timetable has been entered into IS. T. Pavlík
Course Enrolment Limitations
The course is offered to students of any study field.
Course objectives
Advanced biostatistics concepts heavily relying on computing and algorithms as well as selected approaches are presented in this course and supplemented with typical applications. The classical paradigm of statistical modelling is compared to the more recent paradigm of statistical learning. The students are confronted with basic problems typical for modern bioscience research. Approaches discussed are multiple testing procedures and thresholding alternatives, generalized additive non- and semiparametric binary regression models, penalized binary regression models, and support vector machines. The series of lectures is accompanied by an R-based computer laboratory with practical data analysis and the discussion of case studies as well as selected papers.
Syllabus (in Czech)
  • Basics The characteristics of biostatistics and statistical biocomputing with respect to applied statistics, computational biology and bioinformatics. Design aspects (experimental, quasi-experimental, and non-experimental). Data sources and measurement characteristics (conventional sampling, time sampling, and event sampling). Data structures (errors, complexity, size, and dimensionality) in the modern biosciences. The role of statistical (stochastic) methodology. The role of computing and algorithms. Inferential (frequentistic and Bayes) approaches and their limitations. Statistical modelling and its assumptions (distributions, parametric relationships). Statistical (machine) learning and its (lack of) assumptions. Curse of dimensionality and ill-posed problems. Complexity control, regularization and penalization. Selected approaches and applications Multiple testing procedures and thresholding alternatives. Application to microarray gene expression data (n<
Language of instruction
English
Further Comments
The course is taught only once.
  • Enrolment Statistics (recent)
  • Permalink: https://is.muni.cz/course/sci/autumn2005/MSchim