Klapwijk et al. Sample size estimation for task-related functional MRI studies using Bayesian updating

[eduardklap]

Repository: https://github.com/eduardklap/sample-size-codecheck

Certificate: 2024-005

[nuest]

Hi @eduardklap ! Thanks for opening the PR - can you give me some context on the background, please? Are you working in the context of CHECK-NL or with AUMC?

[eduardklap]

Hi @nuest, thanks for your message! Although I am involved in the organisation of the CHECK-NL workshop on 28 November in Rotterdam, this PR is just because I want to get my own paper checked 😄
I tried to follow the CODECHECK instructions for authors and opened this PR. Would it be possible to codecheck this manuscript?

[nuest]

@foost Just because of spatial proximity - what do you think about approaching one of the recently added people from CHECK-NL events? See the last few lines at https://github.com/codecheckers/codecheckers/blob/master/codecheckers.csv

[eduardklap]

Hi both, just FYI in principle everyone with RStudio/Quarto installed and a bit of computing power at hand should be able to rerun the analyses (at least, that is what I aimed for 😄), maybe that helps in finding someone to do the checking

[nuest]

@LukasRoeseler would you be available to conduct a community CODECHECK here for a preprint?

Since this is your first one, I'd follow along a bit closer, we can also have a short call at the start and towards the end of your reproduction. Apart from that, everything you need to know should be in https://codecheck.org.uk/guide/community-workflow (and we're always looking for feedback).

@eduardklap Did you submit the paper already? Would you be able or interested to reference the check in the paper?

[eduardklap]

I did submit the paper and I would love to reference the check in the paper. Revision date is due beginning of October, so it would be great if it can be checked in the meantime. Thanks in advance!

[LukasRoeseler]

I am unavailable this week and next week, so I may not be able to make it until beginning of October. If this is not a problem, I will gladly accept the codecheck!

Also for clarity, I have not attended the CHECK-NL events.

[eduardklap]

Thanks Lukas! I think I will manage to squeeze it in the paper (will already add a sentence about the CODECHECK, can later add the doi if everything goes well)
Looking forward to your findings!

[nuest]

Excellent!

Thanks @LukasRoeseler - let me know when you have taken a look at the documentation, and then we can schedule a short call so I can help you get started.

[eduardklap]

Hi folks, just wanted to check if you think it is feasible to do the codecheck in the near future?
Would be great since I am finishing up the revision. Thanks in advance!

[LukasRoeseler]

Hi, Daniel and I met today and I am ready to start now. I will try to get this done this week. I am very sorry for the long delay!

[eduardklap]

No worries! That sounds great.
Let me know if there is anything unclear about the code

[LukasRoeseler]

I could reproduce all figures successfully and I am now writing up the report.

There are currently two problems. I apologise in case some of this is basic knowledge about github that I am lacking:

1. I cannot push my codecheck files to the repository. @eduardklap can you please grant me permission to upload the files?
   This is the error that I get when trying to push my commits:

remote: Permission to eduardklap/sample-size-codecheck.git denied to LukasRoeseler.
fatal: unable to access 'https://github.com/eduardklap/sample-size-codecheck.git/': The requested URL returned error: 403

2. There is a problem with the yaml file, can you please assist? Tagging @nuest as you probably know the most about the code.

codecheck_yaml = yaml::read_yaml(here::here("codecheck.yml")) (taken from this example) returns the following error:

Error in yaml.load(string, error.label = error.label, ...) :
(.../sample-size-codecheck/codecheck.yml) Scanner error: mapping values are not allowed in this context at line 20, column 12

I have installed the "here" and "yaml" packages and cannot figure out what the problem is. I compared the present .yml file with this one but could not find any reason why the code fails to run.

[eduardklap]

Great to hear that, thanks so much!

1. I think the best way to go is to first fork the repository, then you are able to make the changes in your own fork. Checking with @nuest if this should indeed be the workflow?
2. Thanks for checking this, there was a typo in the yaml file. I just corrected it and pushed the updated file to the repo: now it should be fine

[nuest]

Yes, that's possible.

@LukasRoeseler Please fork into the codecheckers organisation, you should have the rights to do that.

@eduardklap It would be great though if you accept the subsequent PR to update the CODECHECK configuration file, only then can we use this repository as the "checked repository", because that is where the normative metadata record should be.

@LukasRoeseler This is probably something we can improve in the workflow. In case the author does not react, we can use our fork. In this case, I think we can update the author's repo though. The fork is also useful because an author may just completely overhaul their project or even remove it, while we keep our clone "as is" at the time of the check. IIRC the workflow should tell you to fork and eventually also "archive" the fork.

[LukasRoeseler]

I finished the codecheck. The zenodo preprint is published and the final report is available in the codecheck folder: https://github.com/codecheckers/sample-size-codecheck/blob/main/codecheck/codecheck.pdf

Röseler, L. (2024). CODECHECK certificate 2024-005. Zenodo. https://doi.org/10.5281/zenodo.13945051

[nuest]

@eduardklap Congratulations on a successful check!

It would be great if you can incorporate the check in your submission. We're also always interest in conversations with journals, so if your handling editor picks up on the check feel free to point them to us or maybe you can make a joint conversation happen.

Thank you for your contribution to advancing research towards more open and reproducible research!

[eduardklap]

Great news! Thank you so much for checking @LukasRoeseler and sorry for consuming all those computing hours :)
Thank you @nuest for organizing the whole check so nicely. Will definitely cite the check and the CODECHECK project in the resubmission of the paper. Will also put the editors in contact in case they are interested.

[LukasRoeseler]

You're welcome @eduardklap, thank you for making the check so easy for me! And apologies for taking so long!