Needs Analysis Workshop/Breakout group 1

From phenoscape
Revision as of 20:03, 18 January 2008 by Jpb15 (talk | contribs)
(diff) ← Older revision | Latest revision (diff) | Newer revision → (diff)

Break-out Group 1: Developmental genetics of morphology

Examples for driving research questions:

  • Which genes are known to be expressed in the development of a particular morphological structure?
  • Is there a model organism mutant that has a phenotype similar to a human disease?

Characters

GW - describe natural units - not necessarily arbitrary - can phenotype be decomposed into natural mechanistic units

CK - could ontology hold you back from describing some things

GW - can lead to confusion

EJ - this database can be a way of discovering these units

GW - would database be like GenBank - I could describe a fish and deposit my phenotypes - structure of data would be fluid

PM - data is representation of current knowledge - can change over time

GW - how is change in knowledge, like of orthology statements, handled in genome databases

Manually updated, typically

MW - We should move above mechanics and focus on biological questions

EJ - how can we use data to identify discrete modules of variation within evolution

GW - taxonomic work in mammals often involves tooth characters - usually cusps are individualized - actually character identity should be attached to tooth type rather than cusp as unit character - resolves many homology issues

Question: See same tooth type evolving again and again. Developmentally, cusps don't come and go individually. - see covariance within mutants query for mutants that affect cusp X - always see that other cusp is affected


GW - Homology and analogy - often too "yes or no" - instead could have "character model" - contain kind of cells within, location in body, relation to other characters, what kind of states can it assume - represent variational tendencies of a chunk of the body - avoid simplifying biological reality - open-ended structure

CK - can you really ask question like "what are modules of evolution" from database - is that too big

PM and MW - hope to draw such conclusions from database

EJ - quantify levels of covariation across clade as well as across experimental phenotypes


Question: Paired appendages in vertebrates - fins and limbs etc. Homologous, but endoskeletons are not comparable though Do fish have digits (digit homolog)? Is there a skeletal structure that is historically continuous with digits? Are there mutants in the mouse that affect phalanges? Are there similar zebrafish mutants that affect radials? Position, shape, cell type of skeletal elements? Similar gene expression patterns? Does it support continuity of developmental identity? Evidence that there are intermediate forms - continuity - require fossil data

  1. Are there mutants that affect phalanges?
  2. What genes are involved?
  3. Are there mutants in "similar" structures affected by any of those genes?
    • Point you to best candidates in terms of phenotypic similarity, membership in pathway

Compare in database?

Question: Comparative catalog of cell types

Question: Does antennal structure vary more than legs in beetles? Developmental reasons why? Or function/natural selection? Hypothesis: pathways mediating development of leg is more tightly hardwired than in development of antenna Data needed:

  • score amount of variation across phenotypes, mutant affects, pathways
  • measure phenotypic affect of a mutation on leg vs. antenna - how to measure magnitude of phenotypic change in either structure?
    • current data are biased by screens

Where do patterns of covariation change within evolution?

  • identify using database, give hints to experimental work

See change in developmental affects in genes - do you see similar evolutionary change?

See rapid evolutionary change in gene sequence within phylogeny - what phenotypic changes are associated with that part of phylogeny?

  • investigate possible correlations
  • requires phylogeny
  • map phenotypic changes on phylogeny
    • Generate phylogenies on the fly
    • User submit phylogeny
    • User play with topology on screen and see results

Negative information?

  • Database tell you what the gaps are that would allow you to answer a particular question

Estimate rates of trait evolution?

View synapomorphies for groups

As data is provided by users from field

  • need latitude and longitude for specimen - can immediately interface with ecological data
  • developmental stage of specimen? usually adult, usually give you size
  • from specimen ID, we can cross reference to much of this information

Gene networks

  • are phenotypic changes non-randomly placed within gene networks?
  • hypothesis: can't play around much with receptors, but other components in signaling pathway can vary
  • is particular node within pathway correlated with major phenotypic change

Repository of micro-CT data - slice electronically - sort into series - find by ontological term association

Map where shape changes on phylogeny

  • ask what else changes more than expected along those particular nodes
  • are than any structures changing in other groups in structures considered homologues

Back to module discovery... what questions would you ask?

  • Look for discrete piece of development in developmental series
  • What pieces of development share a piece of genetic program?
  • Are the structures influenced by a common developmental trigger? Even if not contiguous in space...
  • Intermodule connections are biologically less stable than intramodule
  • what is known about signals that co-influence structures?