If you’re interested in paleodiversity I just implemented our TRiPS approach in a web-based app, using ShinyApps.io. It’s rather straightforward. The app has a sliding window to select a temporal interval for analyses to be done, you input a clade name and the app automatically downloads the relevant data from PBDB, analyzes it and outputs the results in both table and figure format.
The analysis is performed for each geological stage, which are also downloaded from PBDB. Only observations that are coded as ‘regular’ taxa are used, i.e. oo- and ichno-taxa are ignored, and if the temporal placement of an observation spans more than 2 stages it is not used. Also note that this is analysis at the level of species, so only observations that have accepted species names are used (in other words, fossils identified to genus or higher levels are not used).
There is also a tab to see the observation matrix used, i.e. how many times a particular species was observed within the different intervals. This can be used to further vet the data, people with in-depth knowledge on the taxonomy of particular clades will probably find some errors here that warrant corrections.
There is also a setting for how the maximum number of observations to be downloaded. This is there to speed up the app, since downloading a large number of observations both takes time on the PBDB server and makes analysis more time consuming. Note that if there are more observations on PBDB than sent to the app, the results will often look weird for the later stages, since PBDB outputs them in a particular non-random manner. If the statement on how many observations included in the analysis (on the Occurrence Matrix tab) is close to the set maximum number, then all observations in PBDB were probably not used, so increase this number.
The people doing the actual entering of data into the Paleobiology Database deserves credit for their work. The last tab in the app (PBDB enterers) shows how many observations each of the different contributors have coded into the PBDB. Be sure to thank them. They deserve it.
For more details on the methodology, you can read the original paper here. If you have any comments to this please drop me an email or comment below.
Jostein Starrfelt 