Bioinformatics network giving a new value to agriculture

1
Bioinformatics network -giving a new value to
agriculture

• Irina Arhipova, Liga Paura, Latvia University of
  Agriculture
• Tonu Kollo, University of Tartu
• Ilona Miceikiene, Lithuanian Veterinary Academy
• Lars Snipen, Norwegian University of Life Science

2
IT and biology?

• What is the result of teaching biology to
  information technology students, and information
  technology to biologists?
• Do we get out specialists in bioinformatics?

3
What is bioinformatics?

• Bioinformatics is the interface between computer
  science and molecular biology through statistical
  data analysis.
• With the strengthening IT facilities for
  agricultural research in the Baltic countries, it
  was inevitable that the subject of Bioinformatics
  would arise.

4
Bioinformatics approaches

• Bioinformatics can be approached from two
  sides
• Either biologists learn information technology
  (IT) to use within their specialty, or
• IT specialists learn biology to apply their
  skills to solve biological problems.

5
The history of network development

• The first study course in bioinformatics was held
  in Jelgava, Latvia University of Agriculture in
  2004 under the BOVA network.
• This was the initial step to establish
  international network on the topic in 2005 under
  the Nordplus Neighbour programme.

6
The history of network development

• NOVA-BOVA Bioinformatics 6 ECTS4CP course in
  22.-27 November, 2004
• Teachers Peter Sestoft, KVL, Juris Viksna,
  Latvia University.
• 3 MSc students and 16 PhD students from Latvia,
  Lithuania and Estonia.
• Bioinformatics network meeting in Kaunas,
  Lithuanian Veterinary Academy, December, 2004.

7
Nordplus Neighbour 2005-2007 program project
Nordic-Baltic-Russian Academic Network in
Bioinformatics

• Latvia University of Agriculture, Faculty of
  Information Technologies,
• Royal Veterinary and Agricultural University,
• Norwegian University of Life Sciences,
• University of Tartu, Faculty of Mathematics and
  Computer Science,
• Lithuanian University of Agriculture, Institute
  of Information Technologies,
• Lithuanian Veterinary Academy,
• Saint-Petersburg State Agrarian University,
• Institute of Farm Animal Genetics and Breeding
  Russian Academy of Agricultural Science.

8
The aim of the cooperation

• The development of Nordic-Baltic-Russian network
  in the bioinformatics field, to promote higher
  education study courses or programs for MSc
  students.
• Its proposed effect is the development of 2-years
  joint MSc programme and related research in the
  field of bioinformatics.

9
2005/2006 academic year

• Network meeting in Kaunas, Lithuania, January,
  2006
• Biology course content, teaching staff and
  student group identification,
• project application for 2006/07 development.
• Biology course in 21 28 May, 2006, Lithuania.
• Network meeting in Aas, Skiphelle, Norway June,
  2006
• evaluation of the network activities,
  confirmation of the network plan for 2006/2007,
• Biological data analysis, Bioinformatics course
  content.

10
Biology course

• Biology course for MSc students, hosted by
  Lithuanian Veterinary Academy in May 2006
• Introduce students with non-biological
  knowledge to some biology.
• Introduce students with biological knowledge to
  advanced biology.

11
Biology course content

• Biological make up living organisms. Cells and
  chromosomes.
• DNA and proteins. DNA code. Central dogma. Genes.
• Cytogenetical analysis of domestic animal
  chromosomes.
• Genetic modifications. Biotechnology. Bioethics.
• Molecular organization of avian genome.
• Molecular testing methods of nucleic acids.
• Comparitive genomics of avian and mammalian
  species.
• Positional cloning of QTL and searching for
  candidate genes.
• Gene expression profiling.
• Biotechnology methods in rapeseed breeding.
• Practical work on data mining.

12
Biology course statistics

• 21 course participants
• 4 from Estonia, 6 from Latvia, 9 from Lithiania
  and 2 from Russia,
• 2 Bsc students, 7 MSc students, 9 Ph.D. students
  and 3 university staff.
• 8 course teachers
• 1 from Latvia,
• 6 from Lithuania,
• 1 from Russia.

13
Biology course evaluation

• Overall evaluation of course is very good (8
  points in scale from 0 to 10)
• 90 evaluated the course topic as very good or
  good.
• 50 evaluated the distance learning part as very
  valuable.
• 81 evaluated the amount of the material as
  adequate.
• 76 evaluated the course and reading materials as
  new and appropriate.
• Testing of pre-course knowledge is recommended.
• At least 50 of practical works and discussions
  are recommended from all course program.
• The main reasons to participate in the course
  were
• Course topics and contents fit my studies.
• Recommendation by my research supervisor or
  teacher.
• Improve my skills in English.

14
2006/2007 academic year

• Network meeting in Tartu, Estonia, October, 2006
• Biology data analysis course content,
• Bioinformatics course content,
• teaching staff and student group identification.
• Biology data analysis course in 29 January 3
  February, 2007, Estonia.
• Bioinformatics course in 7 12 May, 2007,
  Latvia.
• Network meeting in Copenhagen, Denmark, June,
  2007
• evaluation of the network activities,
• final report development.

15
Biological data analysis course

• Biological data analysis for MSc, Ph.D. students,
  hosted by Tartu University, Estonia, in
  January-February 2007.
• The aim of the "Biological Data Analysis" course
  is to provide students with basic knowledge and
  elementary skills needed for the statistical
  analysis of biological data using statistical
  software R.
• The course consisted of the distance learning
  part in November December, 2006 before the
  course and face-to-face part.

16
Biological data analysis course content (1)

• Population and sample, sampling variability,
  normal distribution.
• Prediction interval confidence interval.
• Hypothesis testing. Type I and Type II Errors
  confidence level p-value.
• t-test (hypothesis about expectation, paired
  samples, unpaired samples), chi-square test.
• Statistical dependence (association). Linear
  regression model, principles of estimation,
  determination and correlation coefficients.
• Multiple regression, model building. Testing
  goodness-of-fit.
• Modeling non-linear relationships with linear
  regression.

17
Biological data analysis course content (2)

• Indicator variables. 1-way ANOVA. Different
  parameterizations.
• Issues of multiple testing. Bonferroni
  method/Tukey HSD.
• 2-way ANOVA, Analysis of Covariance.
  Interactions, model assumptions, checking model
  assumptions. Transformations to achieve
  normality.
• Causality. Randomized experiments vs
  observational data. Basic experimental designs.
• Hierarchical clustering, k-means. Estimating the
  number of clusters.
• Principal Component Analysis

18
Biology data analysis course statistics

• 26 course participants
• 9 from Estonia, 12 from Latvia and 5 from
  Lithuania,
• 7 MSc students, 17 Ph.D. students and 2
  university staff.
• 7 course teachers
• 1 from Latvia,
• 5 from Estonia,
• 1 from Norway.

19
Biology data analysis course evaluation

• Overall evaluation of course is very good (8,35
  points in scale from 0 to 10)
• 100 evaluated the course topic as very good or
  good.
• 25 evaluated the distance learning part as very
  valuable.
• 95 evaluated the amount of the material as
  adequate.
• 90 evaluated the course and reading materials as
  new and appropriate.
• Testing of pre-course knowledge is recommended.
• The main reasons to participate in the course
  were
• Course topics and contents fit my studies.
• Recommendation by my research supervisor or
  teacher.
• Improve my skills in English.

20
Bioinformatics course

• Bioinformatics for MSc, Ph.D. students, hosted by
  Latvia University of Agriculture, Latvia, in May
  2007.
• The aim of the Bioinformatics course is to
  provide students with basic knowledge in sequence
  analysis, phylogeny and analysis of microarray
  data, and how to use appropriate software
  available on the internet.
• the distance learning part in April, 2007 before
  the course and the face-to-face part.

21
Bioinformatics course content

• Sequence analysis, pairwise alignments, BLAST,
  PSI-BLAST.
• Phylogeny.
• Microarray technologies, single color and two
  colors arrays, image analysis and preprocessing
  of microarray data, clustering.
• Identifying differentially expressed genes, false
  discovery rate (FDR).
• Experimental designs, weighted analysis of
  microarray experiments.

22
Bioinformatics course statistics

• 10 course participants
• 4 from Estonia, 5 from Latvia and 1 from
  Lithuania,
• 2 MSc students, 5 Ph.D. students and 3 university
  staff.
• 6 course teachers
• 3 from Latvia,
• 2 from Denmark,
• 1 from Norway.

23
Bioinformatics course evaluation

• Overall evaluation of course is very good (7,56
  points in scale from 0 to 10)
• 90 evaluated the course topic as very good or
  good.
• 20 evaluated the distance learning part as very
  valuable.
• 100 evaluated the amount of the material as
  adequate.
• 85 evaluated the course and reading materials as
  new and appropriate.
• The main reasons to participate in the course
  were
• Course topics and contents fit my studies.
• To meet colleagues from other countries and
  universities
• Because of participation of teachers from Nordic
  universities

24
Network meeting in Denmark

• Evaluation of the network activities and final
  report preparation.
• Discussion about MSc study program in
  Bioinformatics development.
• Plans for future cooperation.

25
Master study program Information technologies
(120 ECTS)

• Director of study program prof. Irina Arhipova
• Directors of the study program trends
• Production computer control systems
• Dr.habil.ing., professor Peteris Rivza
• System analysis
• Dr.sc.comp., associate professor R. Cevere
• Information technologies in biosystems
• Dr.agr., associate professor Liga Paura

26
The aim of the master study program

• To prepare the students for the independent
  research in the field of information technologies
  (IT).
• The aim of the trend IT in biosystems of the
  masters study program is to prepare qualified
  and creative specialists, who would be able to
  apply new solutions of IT for storing and
  analyzing biological data.
• Degree to obtain Masters degree of engineering
  in the information technologies

27
Study plan

• The studies of the theoretical aspects of
  information technologies sub-branch (45 ECTS)
• 1.1. The general courses of specialty (Part A),24
  ECTS.
• 1.2. Part of compulsory option (Part B), 21 ECTS.
• Approbation of theoretical aspects within topical
  problems of IT (36 ECTS)
• 2.1. Special course (Part A), 27 ECTS
• Production computer control systems.
• System analysis.
• IT in biological systems.
• 2.2. Optional part (Part B), 9 ECTS.
• Free option (Part C), 9 ECTS
• Development and defence of masters paper, 30
  ECTS.

28
Special course (Part A), 27 ECTS IT in
biological systems

• Introduction to Bioinformatics 4,5 ECTS.
• Data structure and algorithms in Bioinformatics
  I 6 ECTS.
• Data structure and algorithms in Bioinformatics
  II 4,5 ECTS.
• Modelling of Biosystems 4,5 ECTS.
• Analysis of Control in Cellular Processes 3
  ECTS.
• Multivariate biodata analysis 4,5 ECTS.

29
Introduction to Bioinformatics

• Introduction into molecular biology.
• The basic part of the course is centered around
• the problems of analysis of biological sequences
  (protein, RNA and DNA) and
• the corresponding solutions for these problems,
• Introduction to analysis of protein structures
  and phylogenetics.
• Main algorithmic problems and their solutions.

30
Data structure and algorithms in Bioinformatics I

• Students get introduced in most important
  algorithmic methods used within the field.
• Course emphasizes bioinformatics problems that
  are most important with respect to practical
  applications - protein and nucleotide sequence
  and protein structure analysis.
• Course also gives a brief introduction in main
  bioinformatics databases, their dependencies and
  uses.

31
Data structure and algorithms in Bioinformatics II

• Advanced problems of bioinformatics and methods
  and algorithms for their solution.
• Emphasized are the areas of bioinformatics, which
  already now offers solutions that are useful in
  practice
• phylogenetic trees,
• genomics,
• applications of machine learning and
• data mining in bioinformatics.
• Students learn about algorithms for the solution
  of these problems, and about available programs
  and databases.

32
Modeling of biosystems

• Living organisms in biosphere, medicine and
  industry. Common features of biological objects.
  Inheritance. Genome.
• Main classes of biomolecules and their functions
  structural elements, transport systems,
  biocatalysers, transformers of energy,
  regulators, carriers of information.
• Steady states of biological systems. Different
  life cycles of biological and technical systems.
• Cybernetics. Basics of structural modelling.
• Topological modelling, cycles in topological
  model. Implementation of functional relationships
  into the model. Balance equations.
• Expert survey and their applications. Validation
  and tuning of model parameters.
• Modeling of ecosystems and biological processes
  in industry.

33
Analysis of Control in Cellular Processes

• Cell as a structural element of living organisms.
• Application of cellular control processes in
  different branches of biotechnology, medicine,
  veterinary medicine, food technology, ecology,
  bioenergetics.
• Basics of structural biology of cell.
  Signalization and metabolic networks in space and
  time. Stability. Multistability. Control loops.
  Positive and negative feedback.
• Iterative scientific approach cycle of
  experiment and modeling. Phases of process
  modeling structural model, functional model,
  validation.
• Data bases of cellular molecular processes.
• Systems Biology Markup Language (SBML) standard
  for modeling of bioprocesses. SBML compatible
  simulation software COPASI, CellDesigner, MatLab
  and others.
• Methods of description and analysis of bioprocess
  control.

34
Multivariate biodata analysis

• The aim of this course is to provide students
  with multivariate statistical methods, whereas
  most software in biology and agricultural science
  are based on statistical models.
• The course subjects are analytics methods of data
  classification, the fundamentals of data
  analysis, principal component analysis, factor
  analysis, cluster analysis, discriminant analysis
  and multivariate analysis of variance.

35
Conclusions (1)

• Today, when the amount of biological data is
  increasing at explosive rate, the processing of
  these data is needed.
• It is extremely valuable if the agricultural
  specialists and biologists would be able to
  process and exploit these data.

36
Conclusions (2)

• Concepts from computer science, discrete
  mathematics and statistics are being used
  increasingly to study and describe biological
  systems.
• Therefore teaching computer science at
  agriculture related university should exploit the
  close relationship with mathematics, statistics
  and biology.

37
Conclusions (3)

• Linking courses of biology and computer science
  requires close collaboration of the teaching
  staff from different departments of university
  and is the precondition of the interdisciplinary
  research.
• It is the efficient way to increase the
  communication and future research between
  students today and scientists tomorrow.

38

• Thank you for your attention !