Teaching code

4S004550

Credits

12

Coordinator

Zsuzsanna Liptak

Language

English

Also offered in courses:

• Algorithms for computational biology of the course Master's degree in Molecular and Medical Biotechnology

Scientific Disciplinary Sector (SSD)

INF/01 - INFORMATICS

Courses Single

Authorized

Moodle Seminars 0

The teaching is organized as follows:

Learning objectives

Students will acquire a wealth of advanced analytic tools which constitute the foundational basis of the algorithmic solution of important problems in bioinformatics Knowledge and understanding The aim of the course is to provide the student with the necessary skills and know-how for the design and analysis of algorithmic solutions to fundamental bioinformatics problems. Applying knowledge and understanding The students will acquire the ability to design algorithmic solutions for typical problems in bioinformatics and computational biology, e.g., analysis of “omics”-data. Making judgements The students will be able to identify the critical structural elements of a problem and the most appropriate approaches to tackle complex problems in bioinformatics. Communication The students will acquire the ability to describe with appropriate precision and clarity, to both experts and non-specialists: a bioinformatics problem, its mathematical model and the corresponding solution. Lifelong learning skills The students will be able to deepen their know-how in bioinformatics autonomously. Based on the topics studied and the knowledge acquired, they will be able to read, understand, and apply material from advanced text-books and scientific article.

Prerequisites and basic notions

basic knowledge of discrete mathematics

Program

------------------------
UL: Algorithm design
------------------------
1. Fundamental notions of algorithmic analysis and complexity: Brief recap on graph traversals; shortest path problem; minimum spanning tree algorithms; elements of computational complexity and NP-completeness
2. Models for Genome Rearrangement: (i) approximation algorithms for reversal distance model (sorting unsigned permutations); (ii) the Doble Cut and Join model; (iii) Synteny Distance approximation algorithms
3. Models for DNA assembly: (i) The Shortest Common Superstring problem (SCS), connections to maximum cost TSP, approximation of the maximum compression via weighted matching; (ii) efficient algorithms for the Eulerian path and Eulerian cycle problem.
4. Introduction to data structures for genomic sequences: (i) Basics of Suffix trees and Suffix arrays; (ii) some applications.
------------------------
UL: Bioinformatics algorithms
------------------------
1. Pairwise Sequence Comparison:
(i) Pairwise sequence alignment (global, local) (ii) variants: optimal alignment in linear space, semiglobal, affine gap penalties, (iii) Pairwise alignment in practice: dotplots, BLAST, Scoring matrices
2. String distance measures: (i) edit distance, (ii) LCS distance, (iii) q-gram distance
3. de Bruijn graphs: (i) de Bruijn graphs and de Bruijn sequences, (iii) sequence assembly based on de Bruijn graphs
4. Multiple sequence alignment:
(i) exact DP algorithm, (ii) approximation algorithms, heuristics
5. Phylogenetic reconstruction:
(i) distance based data: ultrametric trees and UPGMA, (ii) distance based data: additive trees and Neighbor Joining (iii) character based data: Perfect phylogeny (PP); (iv) character based data: Small Parsimony, Fitch' algorithm (v) heuristics for Large Parsimony.

Bibliography

Didactic methods

------------------------
UL: Algorithm design
------------------------
Lectures (blackboard and slides)
------------------------
UL: Bioinformatics algorithms
------------------------
lectures as well as exercise sessions; homework which will be discussed in class

Learning assessment procedures

------------------------
UL: Algorithm design
------------------------
To pass the exam, it is necessary to take a written test, consisting of open questions and/or multiple choice questions. The exercises are meant to evaluate the student's knowledge of classical algorithms and analysis tools as seen during the course, as well as their ability to model "new" toy problems and design and analyse algorithmic solutions for it.
------------------------
UL: Bioinformatics algorithms
------------------------
To pass the exam, it is necessary to take a written test. Students who have a grade of over 25 in the written test have to take an additional oral exam.
The written exam consists of theoretical questions (problems studied, analysis of algorithms studied, mathematical properties, which algorithms exist for a problem etc.), as well as applications of algorithms to concrete examples (computing a pairwise alignment with the DP algorithm etc.) In the oral exam, the student will explain in detail their solutions to the written exam, and show to what extent they have mastered the topics.
Students of the Masters in Molecular and medical biotechnology will have separate questions.
(The exam is the same for students who follow the course during the semester and those who do not: frequentanti e no.)

Evaluation criteria

------------------------
UL: Algorithm design
------------------------
Ability to design and analyze discrete models for problems in bioinformatics
------------------------
UL: Bioinformatics algorithms
------------------------
ability to apply the algorithms studied on small examples; ability to explain them formally and correctly; ability to analyze them correctly; ability to choose the correct algorithm; understanding of the context (e.g. complexity of problems studied)

Criteria for the composition of the final grade

The overall grade for "Fundamental Algorithms for Bioinformatics" is the average of the grades for the two modules. The exam is the same for students who follow the course during the semester and those who do not.

Exam language

English