Difference between revisions of "Needs Analysis Workshop/Chalk talks"
From phenoscape
Paula Mabee (talk | contribs) (→Notes from chalk talks:) |
|||
Line 53: | Line 53: | ||
** classify phenotypic variation (pigment, morphology (e.g. tail length), behavior) | ** classify phenotypic variation (pigment, morphology (e.g. tail length), behavior) | ||
** Peromyscus trait evolution | ** Peromyscus trait evolution | ||
+ | |||
+ | [[Category:Needs Analysis Workshop]] |
Latest revision as of 20:03, 18 January 2008
Notes from chalk talks:
- John Lundberg:
- ~30,000 teleost species
- ostariophysi most important group of freshwater fish
- Arhat Abzhanov:
- works on craniofacial evolution
- drivers and evolution of cranial development
- variation; beak development and diversification (morphs) among Galapagos finches
- comparative relationships of developmental pathways of interacting and regulatory genes
- Hans Hofman:
- individual variation in gene expression, pathways, physiology, anatomy, behaviour, etc
- variation at the population and species level, such as ecology, biogeography, natural history, etc
- how to link these two through, for example, a data repository
- how to implement the ability for open-ended queries (when we don't know what questions might be)
- how to connect this to development time, plasticity, genotype/environment interactions
- Toby Kellogg:
- grasses have about 10,000 species, e.g., rice, maize
- which meristem-specific genes have been conserved or evolutionarily changed
- challenge: how to get from a list of candidate genes (involved in floresence) to the 3'UTR to use for in-situs for same or different species?
- designing primers for an entire clade based on a multiple alignment
- automating the pipeline is desire and major challenge
- Austin Mast:
- MorphBank
- Images provide greater information than what we can put into words
- Slide-server interface
- bulk annotation, 'tile' layout
- Elizabeth Jockusch:
- beetle devo-evo; antennae divided into: scape (1), pedicel (1), flagellum (typically 9 segments, but many exceptions)
- Tribolium: flagellum -> funicle and club, with variable number of segments
- candidate genes; using RNAi to induce changes (phenotypes) in 5 sp of Tribolium
- e.g. reduced number of antenna segments, why? e.g. fusions of antenna segments, deletions, failure to form? (empirical devo Q's)
- how to code these phenotypes so one can ask for broader patterns, non-randomness?
- Chuck Kimmel:
- Important: role of cells in forming skeletal shape
- what are the activities of cells controlled by developmental genes
- devo hypothesis: modular assembly?
- what is the broad nature of shape diversity (of, e.g., certain bones)
- evo hyothesis: a missing module?
- how could a database help with the shape transformation; how to code cells and cell activity into process? Retrieval of hypotheses vs. facts
- Günter Wagner:
- evolution of limbs, gene duplication in teleosts, internal development of human
- following the genes to the taxa they lead to
- comparative cell morphology and cell physiology
- what to compare to in organisms that don't have a particular cell type? e.g. myelinated cells comparing to lampreys (which have only glial cells)
- David Stern (Tuesday):
- parsing out categories of morphological features (morphology, physiology, behavior, etc.); trait differences between Drosophila species
- list of mutations that cause phenotypic differences
- experimental evidence for the role of mutation
- database of evolutionary mutations and their phenotypes fixed through drift or selection (EvolGen)
- Hopi Hoekstra (Tuesday):
- genetic basis for variation
- classify phenotypic variation (pigment, morphology (e.g. tail length), behavior)
- Peromyscus trait evolution