Tools and case studies -> Tools and standards?

[pgleeson]

• Add after Knitr: Use of community standards where appropriate should be encouraged. In computational neuroscience for example, PyNN \cite{Davison2008} and NeuroML \cite{CannonEtAl2014} are widely used formats for making models more accessible and portable across multiple simulators. @article{Davison2008, author = {Davison, A. P. and Bruderle, D. and Eppler, J. and Kremkow, J. and Muller, E. and Pecevski, D. and Perrinet, L. and Yger, P.}, title = {Py{NN}: {A} Common Interface for Neuronal Network Simulators}, journal = {Frontiers in Neuroinformatics}, volume = {2}, pages = {11}, year = {2008} } @article{CannonEtAl2014, AUTHOR={Cannon, Robert C and Gleeson, Padraig and Crook, Sharon and Ganapathy, Gautham and Marin, Boris and Piasini, Eugenio and Silver, R. Angus},
  TITLE={{LEMS}: A language for expressing complex biological models in concise and hierarchical form and its use in underpinning {NeuroML} 2},
  JOURNAL={Frontiers in Neuroinformatics},
  VOLUME={8},
  YEAR={2014},
  NUMBER={79},
  DOI={10.3389/fninf.2014.00079},
  ISSN={1662-5196} }
• Rename section to Tools and standards?
• Why Vogels et al 2011? Some code is on ModelDB, but not enough to reproduce all figures, incl main result in fig4. Has proven difficult to reproduce from description in paper...

[sje30]

Thanks, just added that text, but as it owns section "Standards".

Re: Vogels et al. (2011) I added it as I thought its Supporting Online Materials section was quite thorough, and a good example from the journal "Science" . I was not aware there were difficulties in the reproduction -- how much is missing? Are the authors aware some things are missing?

[pgleeson]

Fine here re "Standards" section.

Vogels et al: yes, plenty of info in supp info, but not enough code on ModelDB to reproduce all figs. This is actually no worse or better than most models on MDB... Think the key issue is that the Auryn simulator that was used for this was a moving target at that stage. It's more stable now, but some students in Volker Steuber's lab had some difficulties getting it running with the current version, but eventually got there.

A better example might be this from Jim Bednar: https://senselab.med.yale.edu/ModelDB/ShowModel.cshtml?model=151951

They claim all the figs from the paper can be reproduced: http://ioam.github.io/topographica/_static/stevens_jn13.html

Excellent use of ipython notebooks: http://ioam.github.io/topographica/_static/stevens_jn13.html#Fig05

[sje30]

Vogels et al: yes, plenty of info in supp info, but not enough code on ModelDB to reproduce all figs. This is actually no worse or better than most models on MDB... Think the key issue is that the Auryn simulator that was used for this was a moving target at that stage. It's more stable now, but some students in Volker Steuber's lab had some difficulties getting it running with the current version, but eventually got there.

okay, thanks for the info. I'm minded to keep it in as I think for its time (2011) it should be regarded as an 'early adopter'.

A better example might be this from Jim Bednar: https://senselab.med.yale.edu/ModelDB/ShowModel.cshtml?model=151951

They claim all the figs from the paper can be reproduced: http://ioam.github.io/topographica/_static/stevens_jn13.html

We already have the Frontiers paper from Jim, which I think is most appropriate.

Stephen

[pgleeson]

Yes, sorry, didn't notice the previous use of Stevens et al. Fine to leave it as is for now and I'll try to think up of another example.