Beiko gen gis2-share
- 1. GenGIS 2 New approaches to understand the geography of our microbial world Rob Beiko
- 2. Donovan Parks Timothy Mankowski Mike Porter Brett O`Donnell
- 3. demo: the GenGIS environment 2-24 GenGIS v1: Parks et al (2009) Genome Res
- 4. GenGIS v1 overview GUI (wxPython) Core Output application Saved image files (C++) Scripting interface (Python, R) Data Map – many formats (GDAL) Samples – CSV Sequences – CSV Trees - Newick Crossing minimization + statistical test Supported platforms: Windows XP, Vista, 7; OS X 10.4, 10.5, 10.6 Open source: Creative Commons Attribution – Share Alike 3.0
- 5. what's new in v2 GUI (wxPython) Core Output application Saved image files (C++) Save / restore Scripting interface sessions (Python, R) Data Map – many formats (GDAL) Samples – CSV Sequences – CSV Python plugins Trees - Newick External files Stability improvements, various things now work properly on the Mac Interface updates (legends, data visualizations) Linear axes analysis
- 6. bringing map data into GenGIS • Maps: – MapMaker (included application) – Digital elevation data (Geobase.ca, NASA Shuttle Topography data, etc.) – Images (.png, .tif, etc.)
- 7. three views of the LineP transect Original data: Jody Wright, Steven Hallam
- 9. clustering based on Canberra beta-diversity
- 10. relative abundance of SUP05
- 11. demo: plugins and R scripts Linear regression of group frequencies Heatmap RPy2 script Original data: 10-29 Costello et al. Science 326:1694-1697
- 12. clustering of fecal samples Female subjects: F1 – F3 Male subjects: M1 – M3 Two sampling methods: - TP - Direct from feces Two time points = 4 samples per individual. Do these samples cluster with each other?
- 14. Wood Buffalo National Park • Canada’s largest National Park • UNESCO World Heritage status (Boreal Forest) • Threatened by encroaching development – Oil Sands mining (Alberta) – Metal mining (NWT) – Hydro-electric dams (Peace River, BC) • Natural resources sustain traditional use by Métis and First Nations peoples Photos: D Baird
- 15. biomonitoring 2.0 what is being collected • Benthic invertebrates (COI, 28S) – kick sample • Water (16S, 18S, 28S) – 1L volume • Soil (16S, COI, ITS, 18S, 28S, RbcL) - cores • Terrestrial arthropods (COI, 28S) – malaise / pitfall traps • All samples replicated 3 times • 5 time points in initial study • Lots of metadata (soil chemistry, flooding, etc.)
- 16. biomonitoring 2.0 replication results – 2010 trial • fjej
- 17. biomonitoring 2.0 sampling progress • August 2011 • Samples collected, starting analysis of sequences • 'traditional' taxonomy where applicable (arthropods si, bacteria no) • June 2012 • Samples collected • Future sampling: August 2012, June – August 2013
- 18. biomonitoring 2.0 our three-year mission (and beyond) • Develop robust sampling techniques for sequence- based biomonitoring • Develop and apply different approaches for assessing biodiversity (taxon-based and taxon- free), and compare their performance on WBNP data • Identify whether “reference conditions” can be established against which future samples can be compared
- 19. call for collaborators • Currently underway: – Combined axis tests (Many trees, one optimal gradient) – Regional tests of diversity – Canonical correlation analysis and related – Bio2.0 analysis • Goals: – Integrate with online data sources – Support more data types (especially vector data) – More plugins!
- 20. the long-term goal Online data sources Analysis: with APIs -Geo gradients Automated dataset -Diversity vs. habitat + generation / -Diversity networks Local data visualization -Functional models
- 22. acknowledgments GenGIS developers (Dal) Donovan Parks Mike Porter LineP (UBC) Timothy Mankowski Jody Wright Brett O'Donnell Steven Hallam Kathryn Dunphy Bio2.0 Sylvia Churcher Mike Porter Mehrdad Hajibabaei (Guelph) Suwen Wang Donald Baird, Wendy Monk (UNB) Harman Clair Brian Golding (McMaster) Greg Smolyn Jeff Shatford (Parks Canada) Stephen Brooks Christian Blouin Jacqueline Whalley (Auckland U Tech)
- 24. New Zealand fungus beetle (Agyrtodes labralis) COI phylogeny Ecological niche modelling suggests Marske et al. Mol Ecol (2009) several glacial refugia, phylogenies Data shown in GenGIS suggest transalpine migration
- 25. map
- 26. locations
- 29. axes test
- 30. body site data
- 32. heatmaps using R