Review Request: Metzner

[ChristophMetzner]

Christoph Metzner

Dear @ReScience/editors,

I request a review for the following replication:

AUTHOR

Original article

Title: Modeling GABA alterations in schizophrenia: a link between impaired inhibition and altered gamma and beta range auditory entrainment
Author(s): Vierling-Claassen, D., Siekmeier, P., Stufflebeam, S., & Kopell, N.
Journal (or Conference): Journal of Neurophysiology
Year: 2008
DOI: 10.1152/jn.00870.2007
PDF: http://jn.physiology.org/content/99/5/2656

Replication

Author(s): Christoph Metzner
Repository: https://github.com/ChristophMetzner/ReScience-submission/tree/Metzner-2017
PDF: https://github.com/ChristophMetzner/ReScience-submission/tree/Metzner-2017/article/Metzner-2017.pdf
Keywords: Neuroscience, Gamma and Beta Oscillations, Auditory Entrainment, Schizophrenia, Python
Language: Python
Domain: Computational Neuroscience

Results

• Article has been fully replicated
• Article has been partially replicated
• Article has not been replicated

Potential reviewers

---

EDITOR

• Editor acknowledgment (@oliviaguest 24 April 2017)
• Reviewer 1 (@aaronshifman 25 April 2017)
• Reviewer 2 (@pietromarchesi 25 April 2017)
• Review 1 decision [accept]
• Review 2 decision [accept]
• Editor decision [accept]

[rougier]

@ChristophMetzner Thanks for your submission, we'll assign an editor soon. Do you have a link to an open-access version of the original article ?

[rougier]

@otizonaizit Could you handle this submission ?

[ChristophMetzner]

@rougier Thank you. Here is a link to the original article http://jn.physiology.org/content/99/5/2656

[otizonaizit]

@rougier : Hi, Nicolas! I won't be able to take care of this in a timely manner for the next month or two, so it's probably better to look for a different editor. If you don't find anyone suitable, please come back to me!

[rougier]

@otizonaizit Thanks for the quick answer. I'll look for another editor.

[aaronshifman]

I might be early: but I'd volunteer as reviewer.

[rougier]

@oliviaguest Can you handle this submission ?

[oliviaguest]

Sure.

[oliviaguest]

@rougier whose responsibility is it to re-compile the pdf to make sure the internal references show up without question marks and without just outputting the labels, etc.?

[rougier]

@oliviaguest The one who asks 😄 More seriously, you can ask author(s) to try to correct for the internal references. Because they know their manuscript much better than anyone else.
Most probably this might be due to trailing whitespaces or a pandoc version mismatch. But if not, we (edtiros) will have to give a closer look.

[oliviaguest]

@ChristophMetzner could you take a look at the above issue regarding internal references, please?

[oliviaguest]

@pietromarchesi got back to me super quickly — so we have both reviewers ready to go! 👍

[aaronshifman]

Thank you @ChristophMetzner, I'll take a look this weekend.

[ChristophMetzner]

@aaronshifman I have not yet addressed the Figures, still working on that.

[ChristophMetzner]

@aaronshifman I have just updated the plotting scripts and notebooks.

[ChristophMetzner]

@aaronshifman Regarding your other comments:

• "What do you mean by topology none (table 1), why is topology separate from connectivity."
  By topology I simply mean the arrangement of cells, i.e. are they arranged in a 2D or 3D space having coordinates. While this would be better called geometry, many articles use the word topology (also reflected in the names some simulators use, e.g. NEST has a topology module which arranges cells in 2D or 3D space). In this network there is no need for cells being arranged in space, so there is no 'topology'. I guess the difference to connectivity is also apparent now.
• "you mention the scaling again, but you cannot verify the original code - have you tried to contact the original authors?"
  I had contacted the authors in the beginning, about sharing their code in general, without getting aresponse. Therefore, I did not contact them again asking about the scaling.
• " I appreciate what you're trying to do with the exploration of discrepancies part - I feel like it needs a bit more explanation and clarification. Are you able to estimate how far off the reported parameters and the parameters you would "need" to match the presented results are?"
  Unfortunately, I don't see a way to estimate how far off I am, given that we have this highly nonlinear system. I also see various other ways by which the discrepancies could be explained. For example, a 'small' change in the weight from I to E cells (an increase) would probably lead to a more pronounced 20Hz component for 40Hz drive in the schizophrenia network (since the total inhibition would be increased, and therefore, more E cells would be suppressed eevry other cycle). However, I think that this would again reduce the 40Hz component in response to 20Hz drive in the control network, similar to the noise exploration. I can explore this further, however, I am not sure what we will gain from that, since there are more possible ways than the one described above. A complete and thorough search of the parameter space, I feel, is a little bit beyond the scope of this replication, since the main features are being reproduced and only the effect being slightly reduced. Let me know what you think. @pietromarchesi What do you think?

[ChristophMetzner]

@aaronshifman @pietromarchesi Now there are most of the time two figures per page (except for Figure 1 and Figures 7 and 8). Should I make 7 and 8 smaller so that they fit on one page?

[pietromarchesi]

@ChristophMetzner Thank you for your last commits and for your comments. Regarding the last point you raise in your comments, on exploring the discrepancies, I agree with you, and, although interesting and welcome, I do not find it strictly necessary to investigate further possible reasons for discrepancy, given that the main results have been replicated.

I think two figures per page is nicer, if they could fit nicely and be readable. I still find it suboptimal to have several pages of figures that completely break off the text, but I don't have an immediate solution for that.

I have one more tiny aesthetic comment: in run_main_simulations.py (and perhaps elsewhere), when lines are broken up then the second part of the line that falls below should be indented.

I will be on holiday and with limited access to internet from next week. My overall recommendation at this point in time is that the replication (with a few smaller edits) should be accepted.

[aaronshifman]

@ChristophMetzner Thanks for the commits. I hope to get back to you within the coming week.

[aaronshifman]

Hello @ChristophMetzner, I apologize for the delay in the review. I appreciate your addressing my comments and I have no further scientific issues.

My only issues are a few typos / omissions

• page 1: schizophrenic patientas
• page 1: loosing much of its simplicity
• page 11: features a re a little bit
• page 2, second equation: S_k = \Sigma_j^n ... : n isn't defined in the text (I assume its over all neurons, just a sentence in the text to define it would help)

I agree with @pietromarchesi that the disjointed figures aren't the best option however I don't have a better option and it seems to be that the style that the figure layout isn't nearly as important as the replication itself.

Given this, I move to accept this paper (conditional on these small editorial changes).

[ChristophMetzner]

@aaronshifman @pietromarchesi I have fixed the typos/omissions and have rearranged the figures and the text and I think that the layout of the article is ok now.

[rougier]

@oliviaguest Anything more needed ?

[oliviaguest]

I do not think so. Will move to publish it now.

[oliviaguest]

Done! 😄