search

trvrb / flux

Integrating influenza antigenic dynamics with molecular evolution
http://bedford.io/papers/bedford-flux/
12 stars 5 forks source link

HI data and code availability #4

Closed trvrb closed 10 years ago

trvrb commented 11 years ago

The authors should be commended for making publicly available all of the HI data that they have assembled (a good scientific practice that sadly has not been adhered to in some previous publications on antigenic cartography). However, they should mention in the text the availability of these data on datadryad - I almost missed it since it is only mentioned on the cover page. They could also consider more clearly specifying where the code relevant to this paper is found in the BEAST repository.

plemey commented 11 years ago

Availability now states "Source code implementing the cartographic models has been made fully available as part of the software package BEAST \cite{BEAST17}, and can be downloaded from its Google code repository (http://code.google.com/p/beast-mcmc/). Incidence data used in this analysis is archived with Dryad (doi:10.5061/dryad.rc515)." Perhaps add "Incidence data AND HI DATA used...", because the availability of HI data through Dryad was only mentioned in the cover letter, right? Perhaps we should also mention in the response that the BEAST implementation consist of different classes contributed to different packages and that pointing out the relevant code in the repository would be complex and of little use to the practitioner. An xml example of how to set up the model and perform inference however seems more useful, and perhaps something worth including in the Dryad repository?

trvrb commented 11 years ago

I agree that XMLs should be included somewhere. I'm not sure whether they should be semantically attached to the paper as supplementary code (which eLife accepts) or as part of the Dryad repository.