https://wiki.phenoscape.org/wg/phenoscape/index.php?title=Needs_Analysis_Workshop/Semantics_Breakout&feed=atom&action=historyNeeds Analysis Workshop/Semantics Breakout - Revision history2024-03-29T09:38:18ZRevision history for this page on the wikiMediaWiki 1.31.10https://wiki.phenoscape.org/wg/phenoscape/index.php?title=Needs_Analysis_Workshop/Semantics_Breakout&diff=2467&oldid=prevJpb15 at 20:04, 18 January 20082008-01-18T20:04:30Z<p></p>
<table class="diff diff-contentalign-left" data-mw="interface">
<col class="diff-marker" />
<col class="diff-content" />
<col class="diff-marker" />
<col class="diff-content" />
<tr class="diff-title" lang="en">
<td colspan="2" style="background-color: #fff; color: #222; text-align: center;">← Older revision</td>
<td colspan="2" style="background-color: #fff; color: #222; text-align: center;">Revision as of 20:04, 18 January 2008</td>
</tr><tr><td colspan="2" class="diff-lineno" id="mw-diff-left-l143" >Line 143:</td>
<td colspan="2" class="diff-lineno">Line 143:</td></tr>
<tr><td class='diff-marker'> </td><td style="background-color: #f8f9fa; color: #222; font-size: 88%; border-style: solid; border-width: 1px 1px 1px 4px; border-radius: 0.33em; border-color: #eaecf0; vertical-align: top; white-space: pre-wrap;"></td><td class='diff-marker'> </td><td style="background-color: #f8f9fa; color: #222; font-size: 88%; border-style: solid; border-width: 1px 1px 1px 4px; border-radius: 0.33em; border-color: #eaecf0; vertical-align: top; white-space: pre-wrap;"></td></tr>
<tr><td class='diff-marker'> </td><td style="background-color: #f8f9fa; color: #222; font-size: 88%; border-style: solid; border-width: 1px 1px 1px 4px; border-radius: 0.33em; border-color: #eaecf0; vertical-align: top; white-space: pre-wrap;"><div>AM – would be interesting if you could query for variation in phenotypes; e.g. want to identify a branch, describe the hairs, trichomes; leaf broad vs. terate – are there other instances of genera in Australia that have both broad and terate leaves -</div></td><td class='diff-marker'> </td><td style="background-color: #f8f9fa; color: #222; font-size: 88%; border-style: solid; border-width: 1px 1px 1px 4px; border-radius: 0.33em; border-color: #eaecf0; vertical-align: top; white-space: pre-wrap;"><div>AM – would be interesting if you could query for variation in phenotypes; e.g. want to identify a branch, describe the hairs, trichomes; leaf broad vs. terate – are there other instances of genera in Australia that have both broad and terate leaves -</div></td></tr>
<tr><td colspan="2"> </td><td class='diff-marker'>+</td><td style="color: #222; font-size: 88%; border-style: solid; border-width: 1px 1px 1px 4px; border-radius: 0.33em; border-color: #a3d3ff; vertical-align: top; white-space: pre-wrap;"><div><ins style="font-weight: bold; text-decoration: none;"></ins></div></td></tr>
<tr><td colspan="2"> </td><td class='diff-marker'>+</td><td style="color: #222; font-size: 88%; border-style: solid; border-width: 1px 1px 1px 4px; border-radius: 0.33em; border-color: #a3d3ff; vertical-align: top; white-space: pre-wrap;"><div><ins style="font-weight: bold; text-decoration: none;">[[Category:Needs Analysis Workshop]]</ins></div></td></tr>
</table>Jpb15https://wiki.phenoscape.org/wg/phenoscape/index.php?title=Needs_Analysis_Workshop/Semantics_Breakout&diff=2275&oldid=prevHilmar at 23:34, 19 September 20072007-09-19T23:34:55Z<p></p>
<p><b>New page</b></p><div>'''Participating''': Toby Kellogg, Hilmar Lapp, John Lundberg, Austin Mast, Peter Midford, David Stern<br />
<br />
<br/><br />
----<br />
<br />
Hilmar – David’s talk – changes characterized – fell into categories<br />
Database is a knowledge base - <br />
<br />
DS - What is ontology for change – words to describe change<br />
PM – develops from?<br />
DS – comparison <br />
PM – How to describe change of position ?<br />
<br />
H – Q formalism - entity, quality – extend to include relational qualities<br />
<br />
e.g. red dot, posterior to, midgut (sp. 1)<br/><br />
red dot, anterior to, midgut (sp. 2)<br />
<br />
posterior and anterior – come from PATO – know that this indicates position<br />
<br />
JL - does this assume that entities are stable during transformation<br />
Does this confuse aspects of the process?<br />
<br />
H – just states relative position<br />
<br />
JL – OK if htat’s all you want<br />
<br />
PM – might be able to infer which moved relative to a third structure<br />
<br />
HL – might get data on a mutant that indicate that the red dot changed position<br />
<br />
DS – suppose species diff is posterior/anterior, mutant moves half way – Could you retrieve direction of movement and then retrieve it<br />
<br />
HL – key question – what do we need to be able to infer –<br />
<br />
DS – if you map on phylogeny infer that red dot moves anterior, mutant moves red dot <br />
<br />
AM – would it be less ambiguous if red dot is relative to axis<br />
<br />
DS – if each entity can have multiple quality descriptors, can describe multiple phenomena<br />
<br />
JL – if things are in different positions, need a standard position. Hard to interpret if you are just going through the literature<br />
<br />
HL – need to be very careful of hypothesizing cause<br />
<br />
JL – also interpreting the literature – <br />
<br />
DS – similar to people describing morphometric of wing vs. outline<br />
<br />
PM – would be nice to have a representation that let you see data in two different ways<br />
<br />
DS – would it be better to make database to describe phenotype of each species<br />
<br />
JL – more what we’re doing, descriptions of species, then building up<br />
<br />
PM – if you have developmental data, better to describe in terms of own development<br />
<br />
DS – discover mutations based on distinction between two things – white cow vs. black cow – wt vs. mutant<br />
<br />
TK – a different way to represent data<br />
<br />
DS – not by mutation but rather by alleles – describe data in terms of alleles, not mutation<br />
<br />
PM – need the alleles – do you need a step matrix to indicate if you need <br />
<br />
JL – lots of different colored cows, multiple species, allele information – if you are interested in looking in evolutionary context, get a tree for cows and map color on tree. You would reconstruct steps according to what you know about mutation<br />
<br />
PM – not necessarily a tree, could be a lattice<br />
<br />
JL – this is what we’re trying to do with ZFIN – mutants of zebrafish vs. morphological variation in species of fish<br />
<br />
PM – not trying to infer anything, trying to develop hypotheses<br />
<br />
DS – would be interesting if you could include QTL data – <br />
<br />
TK – so if db were constructed with alleles as entities, then different states of QTL would be the entities<br />
<br />
DS – probably easier to ignore the QTL literature for now and wait until people get the genes<br />
<br />
DS – how to integrate with fish db<br />
<br />
HL – would be useful, but best to use the same formalism<br />
<br />
DS – can you make this in FileMaker and export to your system?<br />
<br />
HL – subject needs to come from an ontology, predicate comes from PATO, relational quality needs to be formalized<br />
<br />
DS – if things are not in ontologies now?<br />
<br />
PM – ask them to be added<br />
<br />
DS – is your work going to be integrated with NCBI? Hoping to have direct links in that direction<br />
<br />
HL – link to NCBI would come through ZFIN; are describing mutant phenotypes of which they know the genetic basis<br />
<br />
JL – for most of the species that we are looking at, there is no information on the genetic basis – expect that genes will be zebrafish-like<br />
<br />
HL – NCBI is expanding to recording phenotypes – have a new database related to clinical studies database, primarily driven through medical applications. One question I was wondering abou in our effort to annotate phenotypes – what is level of completeness for annotations. Some property may be recorded of red dot, then find a specimen that differs in some way – get disconnected phenotypes<br />
<br />
TK – this is inherent in the data <br />
<br />
JL – can’t get a description complete enough to relate to everything to everything – if you have the luxury of being thorough you could go back and score everything – we won’t have the luxury – will be holes in the data<br />
<br />
DS – in this example, cells differ in ability to process salt – would have a separate setoff qualities – only way to deal with it need to have real taxonomic experts dealing with each group – <br />
<br />
JL – some things are very straightforward like numbers of fingers and toes, other things such as relative sizes of parts are harder<br />
<br />
DS – flybase doesn’t have all information translated into ontologies<br />
<br />
JL – trying to seize on all knowledge built up in evolutionary morphology<br />
<br />
DS – is there an intermediate state where sentence can be lifted from papers<br />
<br />
Very detailed ontologies may not be necessary – the people using these may want to look at the organisms themselves<br />
<br />
HL – how would biologists want to take advantage of this in a way that does take advantage of the semantics<br />
<br />
DS – go to flybase – skip the ontology, scan for key words, go and read the paper<br />
<br />
HL – look at your own database, what kinds of queries do you use?<br />
<br />
DS – need to have all data in one place in one format – <br />
<br />
PM – if you are looking for a single word, may be able to scan the text – if you want to look for a term and its relation to something else, ontology might be useful. May want to give me everything in which dot position changes<br />
<br />
HL – relative position – can go higher and higher level<br />
<br />
DS – are we finding only a subset of genes that coule affect this trait? Would like to find a set of genes that could affect a trait. What are all genes that affect trichomes? <br />
<br />
TK – many people just want a list<br />
<br />
JL – if the database becomes fairly dense, could imagine someone using it to help reconstruct a phylogeny – depends on how dense the product is<br />
<br />
TK – possibly the ability to manipulate the data is ahead of the availability of data<br />
<br />
JL – as taxonomic ontology develops and as ontology becomes richer and richer then become more useful<br />
<br />
HL – this project could evolve in to the next BLAST as a way of analyzing phenotypes, similar to that of analyzing sequences. At some point might use images and their annotation as the basis of a BLAST search<br />
<br />
AM – was thinking more about annotations to images – sifting through EQ statements. Would want to search across data sources, not just image annotations<br />
<br />
HL – is it interesting to search for similar phenotypes – is there meaning in doing that?<br />
<br />
AM – would be interesting if you could query for variation in phenotypes; e.g. want to identify a branch, describe the hairs, trichomes; leaf broad vs. terate – are there other instances of genera in Australia that have both broad and terate leaves -</div>Hilmar