search

greenelab / deep-review

A collaboratively written review paper on deep learning, genomics, and precision medicine
https://greenelab.github.io/deep-review/
Other
1.25k stars 270 forks source link

Call for Methylation Section #942

Closed jlevy44 closed 4 years ago

jlevy44 commented 5 years ago

I’ve been reading through the deep review paper and noticed that there is not a Methylation section in this paper. There are a number of issues that have been brought up regarding the inclusion of methylation deep learning papers (#68 #39 #244 #923 #922 #910 #692 ). The only paper I see explicitly mentioned in the paper for deep learning exclusively for methylation data types is DeepCpG (I’ve read the referenced multimodal approaches). However, it seems like there are quite a few papers that demonstrate significant strides in Deep Learning and methylation, most of which are enumerated below:

I believe that the inclusion of methylation in the deep learning paper has been limited, and propose that we include a Methylation section of the paper under the “Deep learning to study the fundamental biological processes underlying human disease section”.

Brock Christensen @Christensen-Lab, Alexander Titus @AlexanderTitus, and I have been working together to draft a Methylation section for the deep review paper, which we are revising internally, and would be happy to submit a pull request soon. The three of us would also like to be included as authors in the paper.

evancofer commented 5 years ago

We'd love to see your contributions. From what you say, the addition of such a subsection seems apropos.

For the sake of efficiency and this project's focus on open/collaborative writing, I recommend that you submit a public pull requests before you get too far along. This will help publicly document all your contributions, ensure that your additions pass automated checks, and make it easier for others to make comments on your proposed additions (and thereby ensure that you're on the right track in terms of content, formatting, etc).

Authorship isn't decided immediately, since authors are defined as anyone who has met ICMJE authorship standards (defined here) at the time of submission. The "at the time of submission" part is critical, e.g. some contributors to the first release were not listed as authors because they did not indicate their approval of the final manuscript. @cgreene may have more to say about this.

cgreene commented 5 years ago

@evancofer's summary seems on point. Things that we required were participation. This means that if you submit a fully formed PR but it only comes from one account, that provides contribution credit to that person but not the others - who may have contributed behind the scenes but did not in a public, attributable way.

To be an author requires meeting all of the criteria, and this does require approval of the manuscript to be submitted. We attempted to obtain this from everyone, but there were authors who did not respond despite efforts to reach out to them, so we included them in only the acknowledgements.

Christensen-Lab commented 5 years ago

Great, thanks for the quick response. We have already begun drafting content, it appears that the "time of submission" will be June 2019 correct? Thanks

evancofer commented 5 years ago

Yes, June 2019 is correct.

jlevy44 commented 5 years ago

Thanks for your quick responses. Brock, Alexander and I will discuss our next steps as we continue to draft the methylation section.

evancofer commented 5 years ago

Great. I'm excited to see what you all have been working on. It sounds like it would make a great addition!

AlexanderTitus commented 5 years ago

Thanks @evancofer and @cgreene. @jlevy44, @Christensen-Lab and I will circle up about how best to divide and contribute.

evancofer commented 5 years ago

Any updates on this?

jlevy44 commented 5 years ago

Hey there. No updates yet. I'll touch base with @AlexanderTitus and @Christensen-Lab so we can start to make more progress.

jlevy44 commented 5 years ago

I believe we're shooting to have something preliminary by mid-February to early-March.

jlevy44 commented 5 years ago

Just providing another update. We’ll be releasing our section through three separate PRs. I’ll PR some motivational material for why deep learning and methylation, and then @Christensen-Lab and @AlexanderTitus will review and edit text that I wrote surrounding methylation inference, predictions, and latent space, and each will submit a PR for their relevant methylation subsection. Hoping to release the intro soon to have something down and in the process of merging.

jlevy44 commented 5 years ago

Hey everyone, just giving an update. We're just waiting on internal edits before we send the entire section out (may update with a few new papers).

Do you have a deadline of when you'd like for us to have this section in by?

evancofer commented 5 years ago

I don't know if there is a deadline for this section per se. The sooner you can get the internal edits in a PR, the sooner you can get some feedback on it. Based on the last PR, feedback may be a little slow, so it is better to post things sooner than later.

jlevy44 commented 5 years ago

Internal review is progressing, should have that PR soon...

jlevy44 commented 5 years ago

What's the best way to multiple PR one section? Internal reviews are complete. Just adding citations. We'd like to make 2-3 PRs, one for each author.

evancofer commented 5 years ago

I am not sure I understand the question, could you clarify? Either way, I am glad to know it's getting closer to PR stage.

agitter commented 5 years ago

@evancofer the question is how to coordinate multiple related PRs so that each contributor gets credit for their text and there aren't complex merge conflicts.

@jlevy44 as long as the PRs don't edit the same existing text, it should be okay to submit them all at once as separate PRs. The last one submitted can reference the PR numbers of the others to help coordinate review. Does that sound okay @evancofer?

Note that the last methylation section PR struggled to find a qualified reviewer, so that may be a problem here as well. We haven't done it before, but maybe you could recruit an external reviewer (i.e. not an existing author) and we could acknowledge them?

jlevy44 commented 5 years ago

Our team has begun the PR process. You should expect 2 more PRs this week.

Here are the order of the PR sections that our group will be submitting:

  1. Introductions - @Christensen-Lab @Christensen-Lab-Dartmouth
  2. Inference, Imputation, and Prediction - @jlevy44
  3. Latent Space Construction and Conclusions - @AlexanderTitus
evancofer commented 5 years ago

I agree with @agitter that finding qualified reviewers is the real issue here. If they have the time, either @michaelmhoffman or @lanagarmire are probably good choices, who IIRC were involved with the first iteration of the deep review

jlevy44 commented 5 years ago

Great. Is there anything I can do to help move things forward at this stage?

michaelmhoffman commented 5 years ago

Thanks for thinking of me, but given my commitments I need to prioritize contributions that will lead to a new paper.

lanagarmire commented 5 years ago

Happy to help and move the manuscript to move forward as a co-author.

Lana Garmire, PhD Associate Professor (with tenure) Department of Computational Medicine and Bioinformatics School of Medicine University of Michigan groupwebsite: garmiregroup.org

On Sat, May 11, 2019 at 8:09 AM Michael Hoffman notifications@github.com wrote:

Thanks for thinking of me, but given my commitments I need to prioritize contributions that will lead to a new paper.

— You are receiving this because you were mentioned. Reply to this email directly, view it on GitHub https://github.com/greenelab/deep-review/issues/942#issuecomment-491505790, or mute the thread https://github.com/notifications/unsubscribe-auth/ACFEMOY67C5NZ6ZM2FP3DFDPU2ZRFANCNFSM4GIDQ5BQ .

evancofer commented 5 years ago

@jlevy44 I think our efforts right now should focus on continuing to update existing sections, and summarizing and discussing new work using the GitHub issues.

@michaelmhoffman No problem. Thank you for filling us in on the situation!

@lanagarmire We'd love to have your contribution. If you're up to reviewing some of the methylation content, I think there are three PRs open right now that cover the methylation sections:

@cgreene Is there a previous PR that stands out as the "ideal" example of how we should do these PR-based reviews? I think it might be useful for onboarding reviewers, and could be worth mentioning in CONTRIBUTING.md.

cgreene commented 5 years ago

@evancofer : this was a new subsection, and may be a good example to work from:

https://github.com/greenelab/deep-review/pull/629

jlevy44 commented 5 years ago

@lanagarmire We'll be more than happy to have you review our sections.

jlevy44 commented 5 years ago

Thank you for the help!

AlexanderTitus commented 5 years ago

If there is anything we can do to facilitate review, please let us know.

We would love to have this section included in the next iteration of the review released, as it adds to the completeness of the review.

jlevy44 commented 5 years ago

@evancofer @cgreene What would you recommend as the next steps for the completion of this review? In particular, what might be some action steps for @lanagarmire to review our content?

Thanks!

lanagarmire commented 5 years ago

I will have a block of time early next week. Is the draft ready for QC/check?

evancofer commented 5 years ago

@lanagarmire I think their current draft is ready for QC. @jlevy44 any last minute changes?

jlevy44 commented 5 years ago

@lanagarmire @evancofer I think we're all good to go. One of our PRs did not pass through Travis, but should be okay content-wise.

Thanks!

cgreene commented 5 years ago

Do you know why Travis failed? Might be good to clean that up first too. Can't merge until it passes, even if PR approved.

On Wed, Jun 12, 2019, 1:30 PM Joshua Levy notifications@github.com wrote:

@lanagarmire https://github.com/lanagarmire @evancofer https://github.com/evancofer I think we're all good to go. One of our PRs did not pass through Travis, but should be okay content-wise.

Thanks!

— You are receiving this because you were mentioned. Reply to this email directly, view it on GitHub https://github.com/greenelab/deep-review/issues/942?email_source=notifications&email_token=AAEEPM3CD435GIZYETGUZNTP2DM55A5CNFSM4GIDQ5B2YY3PNVWWK3TUL52HS4DFVREXG43VMVBW63LNMVXHJKTDN5WW2ZLOORPWSZGODXQDLLY#issuecomment-501233071, or mute the thread https://github.com/notifications/unsubscribe-auth/AAEEPM4A3KOADVGEGSR73WDP2DM55ANCNFSM4GIDQ5BQ .

agitter commented 5 years ago

@cgreene it looks like a space vs. tab failure in the tags file https://github.com/greenelab/deep-review/pull/956#discussion_r292874898

jlevy44 commented 5 years ago

I thought that there was a mechanism in place to automatically reread the tsv with a new delimiter. If you check the Travis output for my PR, the same error may be embedded?

@brockclarke

jlevy44 commented 5 years ago

@lanagarmire looking forward to your review of our material. Hopefully we can finish this review before the submission of the manuscript. Please let me know what I can do to help move things forward.

lanagarmire commented 5 years ago

Does this qualify me as a co-author? Forgot to ask...need to prioritize my time wisely, obviously. Thanks.

On Thu, Jun 20, 2019, 8:02 PM Joshua Levy notifications@github.com wrote:

@lanagarmire https://github.com/lanagarmire looking forward to your review of our material. Hopefully we can finish this review before the submission of the manuscript. Please let me know what I can do to help move things forward.

— You are receiving this because you were mentioned. Reply to this email directly, view it on GitHub https://github.com/greenelab/deep-review/issues/942?email_source=notifications&email_token=ACFEMO5RUSPHRNQT3EXIO3LP3QLBVA5CNFSM4GIDQ5B2YY3PNVWWK3TUL52HS4DFVREXG43VMVBW63LNMVXHJKTDN5WW2ZLOORPWSZGODYHBVFI#issuecomment-504240789, or mute the thread https://github.com/notifications/unsubscribe-auth/ACFEMO75X4TWPOLY742EMC3P3QLBVANCNFSM4GIDQ5BQ .

cgreene commented 5 years ago

@lanagarmire It will depend on the extent to which your review meets the ICMJE guidelines: https://github.com/greenelab/deep-review/blob/master/CONTRIBUTING.md#authorship

The two that are most in question here (the others are approval/agreement) are:

Substantial contributions to the conception or design of the work; or the acquisition, analysis, or interpretation of data for the work; AND

Drafting the work or revising it critically for important intellectual content; AND

If you make specific suggestions to the text that revise it for critically important content (now far more trackable via github's suggest changes feature, which I would strongly encourage you to use so that your suggestions can be tracked) that get incorporated, and make contributions to the design of the sections (i.e., if you identify substantial changes that improve clarity or identify missing areas/literature that should be covered), that would qualify.

On the other hand, if the changes are primarily copyediting, etc then that wouldn't qualify.

If you are uncomfortable with this, then I wouldn't encourage you to do it and we can find another path. If this is something you are comfortable with, we would like to benefit from your expertise!

cgreene commented 5 years ago

@jlevy44 : Can someone accept @agitter's change on https://github.com/greenelab/deep-review/pull/956

The permissions on the fork are set such that I don't have permission to commit to @brockclarke's branch there.

lanagarmire commented 5 years ago

I already contrbuted to critical suggestions before the writing. That is why I was asked to QC.

But if you don't think that matter then i will pass this one.

On Fri, Jun 21, 2019, 8:07 AM Casey Greene notifications@github.com wrote:

@lanagarmire https://github.com/lanagarmire It will depend on the extent to which your review meets the ICMJE guidelines:

https://github.com/greenelab/deep-review/blob/master/CONTRIBUTING.md#authorship

The two that are most in question here (the others are approval/agreement) are:

Substantial contributions to the conception or design of the work; or the acquisition, analysis, or interpretation of data for the work; AND

Drafting the work or revising it critically for important intellectual content; AND

If you make specific suggestions to the text that revise it for critically important content (now far more trackable via github's suggest changes feature, which I would strongly encourage you to use so that your suggestions can be tracked) that get incorporated, and make contributions to the design of the sections (i.e., if you identify substantial changes that improve clarity or identify missing areas/literature that should be covered), that would qualify.

On the other hand, if the changes are primarily copyediting, etc then that wouldn't qualify.

If you are uncomfortable with this, then I wouldn't encourage you to do it and we can find another path. If this is something you are comfortable with, we would like to benefit from your expertise!

— You are receiving this because you were mentioned. Reply to this email directly, view it on GitHub https://github.com/greenelab/deep-review/issues/942?email_source=notifications&email_token=ACFEMO3NTPEVVVNM3MNW6B3P3TABJA5CNFSM4GIDQ5B2YY3PNVWWK3TUL52HS4DFVREXG43VMVBW63LNMVXHJKTDN5WW2ZLOORPWSZGODYIJC7Y#issuecomment-504402303, or mute the thread https://github.com/notifications/unsubscribe-auth/ACFEMO5BHVGY6JPJ6BF7HLLP3TABJANCNFSM4GIDQ5BQ .

cgreene commented 5 years ago

@lanagarmire : Each contribution matters, but I don't think there would be anyone who could guarantee that ICMJE guidelines for authorship would be met before the review is done. If you feel you need to pass, I totally understand.

lanagarmire commented 5 years ago

Good luck to your review!!!

On Fri, Jun 21, 2019, 10:01 AM Casey Greene notifications@github.com wrote:

@lanagarmire https://github.com/lanagarmire : Each contribution matters, but I don't think there would be anyone who could guarantee that ICMJE guidelines for authorship would be met before the review is done. If you feel you need to pass, I totally understand.

— You are receiving this because you were mentioned. Reply to this email directly, view it on GitHub https://github.com/greenelab/deep-review/issues/942?email_source=notifications&email_token=ACFEMO7X4CCSHYCENEPRLFLP3TNKBA5CNFSM4GIDQ5B2YY3PNVWWK3TUL52HS4DFVREXG43VMVBW63LNMVXHJKTDN5WW2ZLOORPWSZGODYIRP5Y#issuecomment-504436727, or mute the thread https://github.com/notifications/unsubscribe-auth/ACFEMO5W7RXTEECOWBQXB3DP3TNKBANCNFSM4GIDQ5BQ .

jlevy44 commented 5 years ago

@brockclarke Can you accept @agitter ‘s change on #956 ?

@lanagamire Thanks!

@cgreene @agitter @evancofer Is there anyone else you could suggest as a reviewer? I can also ask around.

jlevy44 commented 5 years ago

Hey @cgreene , can you provide us an update with the status of the review of these 3 PRs? Thanks!

Looking forward to moving our contributions to their final state. If we could also add this manuscript to the review before submission that would be great: https://www.biorxiv.org/content/10.1101/692665v1

cgreene commented 5 years ago

Getting this moving again is one of my three main goals of this week! 🤞

jlevy44 commented 5 years ago

Wonderful. Thanks @cgreene !

lanagarmire commented 5 years ago

Please take me off this thread.thx

On Wed, Jul 17, 2019, 7:33 AM Joshua Levy notifications@github.com wrote:

Wonderful. Thanks @cgreene https://github.com/cgreene !

— You are receiving this because you were mentioned. Reply to this email directly, view it on GitHub https://github.com/greenelab/deep-review/issues/942?email_source=notifications&email_token=ACFEMO3ALAWJQCMK4VLJHLDP737QJA5CNFSM4GIDQ5B2YY3PNVWWK3TUL52HS4DFVREXG43VMVBW63LNMVXHJKTDN5WW2ZLOORPWSZGOD2D4WXQ#issuecomment-512215902, or mute the thread https://github.com/notifications/unsubscribe-auth/ACFEMO2BVVDX43V4ZJQ7E4LP737QJANCNFSM4GIDQ5BQ .

cgreene commented 5 years ago

@lanagarmire : your github preferences have you subscribed - nobody else can remove you, but these instructions might be helpful: https://help.github.com/en/articles/subscribing-to-and-unsubscribing-from-notifications

jlevy44 commented 4 years ago

I think we can wrap this up and close this issue once #959 is resolved. Thanks!

agitter commented 4 years ago

Closed by #954 #955 #956

jlevy44 commented 4 years ago

Any interest in having us update this section?