Needs Analysis Workshop/Phylogeny Breakout

From phenoscape
Jump to: navigation, search

Participants: Monte Westerfield, Paula Mabee, Cartik Muthu, Chuck Kimmel, Wasila Dahdul, Hans Hofman

Whiteboard

PM - what are participants most interested in?

-computing tree on the fly?

-repository of trees?

-manipulation of existing trees?

Group consensus - want the ability to deliver trees and ability to manipulate them

HH – want the ability to map as many phenotypic traits as possible;

-character mapping with phenotypic transitions, gains/losses, correlated changes;

-have sets of data mapped on tree – for example , all in situs done for a particular gene…

PM - view changes in all head bones

CK - choose a term and see all changes on tree, perhaps from a drop-down list

Group - discussion of challenges of visualizing multiple characters mapped on the tree

-similar functionality of MacClade, summarizing character change information as tick marks on branches, but this might get crowded

CK – interested in getting figures on the phylogeny

HH - look for correlation between gene expression and trait on tree

HH – note whether in situs have been done at all for species of interest; see which species have that data, download all the images.

CK - interested in output/reports of statistics and quantitative measures, such as statistics for testing character correlation

HH - also independent contrasts to assess character independence

HH - has used heat maps to visualize correlations in data sets; might be useful for this project

CK - time component mapped on phylogeny: overlay movie showing developmental sequence

WD - interest in input of your own data (unpublished) to interact with database?

CK/HH - yes, but challenges of data entry; perhaps a temporary repository for data similar to blast searches

MW - having a modified curator interface available for the user to enter their data and run analyses against the database