Need a way to indicate that taxa are extinct in taxonomy

[josephwb]

I think a lot of taxonomic headaches might be helped if it were possible to specify that taxa are extinct in the taxonomy. For example, one might desire a tree of only extant taxa. Not sure what the criteria for this would be though; Stephen suggested if a taxon appeared only appeared in PaleoDB and no other (because there are extant taxa like Homo sapiens in PaleoDB). Is it possible to add a column to ott2 to include such information?

[pmidford]

Let's make sure we are clear we are distinguishing between extinct (including recently extinct) taxa and taxa only known from fossils. For molecular trees there won't be many of the later, but lack of clarity about that distinction caused some confusion with the Phenoscape taxonomy.

[jar398]

On Mon, Jun 17, 2013 at 10:00 AM, Joseph W. Brown
notifications@github.comwrote:

I think a lot of taxonomic headaches might be helped if it were possible
to specify that taxa are extinct in the taxonomy. For example, one might
desire a tree of only extant taxa. Not sure what the criteria for this
would be though; Stephen suggested if a taxon appeared only appeared in
PaleoDB and no other (because there are extant taxa like Homo sapiens in
PaleoDB). Is it possible to add a column to ott2 to include such
information?

I have been thinking about how to introduce additional GBIF provenance
information into OTT. I could special case PaleoDB but it would be better
to capture the GBIF source field, which currently takes on any of 40
different values. This could conceivably be used for other purposes as
well, e.g. to provide hyperlinks to the original source.

Can you be more specific about what headaches?

Jonathan

—
Reply to this email directly or view it on GitHubhttps://github.com//issues/38
.

[josephwb]

Just that the vast majority of "orphan" taxa (genera, families, etc. with
no species) seem to be extinct ones. To boot, these often attach in weird
places in the taxonomy (like the Paragrallator example I sent earlier,
which is a dinosaur which attaches at the base of Metazoa).

Joseph.

On 17 June 2013 11:06, Jonathan A Rees notifications@github.com wrote:

On Mon, Jun 17, 2013 at 10:00 AM, Joseph W. Brown
notifications@github.comwrote:

I think a lot of taxonomic headaches might be helped if it were possible
to specify that taxa are extinct in the taxonomy. For example, one might
desire a tree of only extant taxa. Not sure what the criteria for this
would be though; Stephen suggested if a taxon appeared only appeared in
PaleoDB and no other (because there are extant taxa like Homo sapiens in
PaleoDB). Is it possible to add a column to ott2 to include such
information?

I have been thinking about how to introduce additional GBIF provenance
information into OTT. I could special case PaleoDB but it would be better
to capture the GBIF source field, which currently takes on any of 40
different values. This could conceivably be used for other purposes as
well, e.g. to provide hyperlinks to the original source.

Can you be more specific about what headaches?

Jonathan

—
Reply to this email directly or view it on GitHub<
https://github.com/OpenTreeOfLife/opentree/issues/38>
.

—
Reply to this email directly or view it on GitHubhttps://github.com//issues/38#issuecomment-19551123
.

[jar398]

I don't know how to implement this pragmatically because the criterion "appears in PaleoDB and no other" is very hard to test. GBIF is made up of 40 taxonomies (and growing fast) and I would have to check all the other 39 - which would mean downloading, parsing, and so on.

[pmidford]

I don't think that would work anyway, there are at least a handful of
extinct taxa (with deposited sequences) in NCBI. These are both recently
extinct (Mammuthus - which also appears in GBIF) and not recently
(Tyrannosaurus).

On Wed, Jun 19, 2013 at 12:55 PM, Jonathan A Rees
notifications@github.comwrote:

I don't know how to implement this pragmatically because the criterion
"appears in PaleoDB and no other" is very hard to test. GBIF is made up of
40 taxonomies (and growing fast) and I would have to check all the other 39

• which would mean downloading, parsing, and so on.

—
Reply to this email directly or view it on GitHubhttps://github.com//issues/38#issuecomment-19697470
.

[jar398]

Just browsing the mammal tree now I've come to agree that this is a pretty serious problem.

IRMNG carries the needed information (extant vs. non-extant) so we should do our own import of IRMNG. Haven't checked paleodb yet.

Issue continued as reference-taxonomy issue 6: OpenTreeOfLife/reference-taxonomy#6

[jar398]

IRMNG is in and we have extinct flags and an extinct() method in the patch system. I think I can close this issue.