R is nowadays probably the most powerful tool for statistics of all types. There are plenty of modules available for work with molecular data. Those will be introduced during the course. Previous knowledge of R is useful, but not necessary. If there is at least one participant not speaking Czech, the course will be in English.
I'd be glad if participants could fill a short questionary which will help me with preparation of the course and communication.
List of topics
- 1st day, morning
- Basic work in R - how to enter commands, install packages, read help, types of variables, indexes, etc.
- This part is not compulsory for participants who already know R, but it is higly recommended as practicing over and over does not hurt. :-)
- 1st day, afternoon
- Load and export molecular data of various types and formats.
- Download molecular data from on-line databases
- Extractions of SNP from sequencing data
- Extraction of polymorphism from sequences
- Mikrosatellites, AFLP, SNP, sequences
- Manipulations with data, conversions among formats
- Distance matrices, import of custom matrices
- Phylogenetic trees (NJ, UPGMA, ML) and display and test
- Basic statistics, genetic indices heterozygosity, HWE, F-statistics
- Export of figures
- 2nd day
- Whole genome SNP data
- Spatial analysis - Mantel test, Moran’s I, Monmonier, sPCA
- Basic map creation
- Manipulations with trees, work with big sets of trees
- 3rd day
- Phylogenetic independent contrast
- Phylogenetic autocorrelation
- Phylogenetic PCA
- Ancestral state reconstruction
- And more...
Used packages: PBSmapping, ParallelStructure, RandomFields, RgoogleMaps, TeachingDemos, XML, ade4, adegenet, adephylo, akima, ape, colorspace, combinat, corrplot, fields, gplots, grid, ips, lattice, mapdata, mapproj, maps, maptools, muscle, pegas, ermute, phangorn, phyloch, phytools, poppr, rworldmap, seqinr, sp, spam, tcltk, vegan.
For course you need
- Working Wi-Fi. Eduroam or in application form You can ask for temporary password.
- Installed R. I also recommend to install some graphical user interface like RStudio, RKWard, R commander or some similar according to your choice.
Changes from last year (based on feedback of participants)
- Updates regarding new versions of R packages
- More theory regarding statistical methods itself
In case of any questions, wishes, comments just ask! Using comment form below, by mail or so on.