search

jaybee84 / ml-in-rd

Manuscript for perspective on machine learning in rare disease
Other
2 stars 1 forks source link

Target Audience of this article #126

Open jaybee84 opened 3 years ago

jaybee84 commented 3 years ago

According to our discussion on 12/04/20, our target audience for this article is:

  1. Anyone who is familiar with the concepts used in ML (e.g. supervised and unsupervised learning, bootstrap, ensemble learning etc), and considers/implements models on large datasets, but dismisses rare disease datasets as too small to be considered for ML. This would include a gradient from someone who has implemented one ML model to a well-known large dataset in a data science bootcamp to someone who regularly thinks about implementing ML in large datasets). Our hope is to urge them to consider using their skills in rare disease while being aware of all the challenges that come with working with a small dataset.
  2. Anyone who is interested/currently works in rare disease, and would benefit from understanding that machine learning can be fruitfully used in their field of study. We hope that this article will help them partner with computational researchers in a meaningful manner.

So, for the text of the article we aim to:

  1. Briefly describe an ML concept (a line or two in simplified terms) so that it makes sense to a reader even if they are not familiar with the ML terminology (ML aficionados will recognize the terminology directly)
  2. Highlight the main challenge in rare disease that necessitates the use of this concept
  3. Describe 1 or 2 examples where this concept was implemented in a rare disease dataset or a rare variant or a small dataset and how it overcame the said challenge.

For the figures of the article, we aim to :

  1. Have a generalized and simplified representation of the concept to engage the ML aficionado as well as the non-ML audience
  2. Incorporate enough technical details as permitted without making the figures hard to understand
  3. Highlight the high-level concepts that necessitate the use of a particular method in a rare disease dataset (aka "add the rare disease flavor")
allaway commented 3 years ago

I see that you assigned this to me, just to clarify, what is the action item? Thanks!

On Fri, Dec 4, 2020 at 12:33 PM Jineta Banerjee notifications@github.com wrote:

Assigned #126 https://github.com/jaybee84/ml-in-rd/issues/126 to @allaway https://github.com/allaway.

— You are receiving this because you were assigned. Reply to this email directly, view it on GitHub https://github.com/jaybee84/ml-in-rd/issues/126#event-4074368099, or unsubscribe https://github.com/notifications/unsubscribe-auth/AE3WNSFKW6TBHCDETSBWJSTSTFBRRANCNFSM4UN5RK4A .

jaybee84 commented 3 years ago

Assigned mostly to get everyone's attention through targeted notification :)

One action item: see directions for article text (paragraph 2) to guide the future edits of the article :)

allaway commented 3 years ago

Ah, haha, I only saw that it was assigned to me via email. makes sense!

On Fri, Dec 4, 2020 at 1:22 PM Jineta Banerjee notifications@github.com wrote:

No Action item here, assigned to get everyone's attention through targeted notification :)

— You are receiving this because you were mentioned. Reply to this email directly, view it on GitHub https://github.com/jaybee84/ml-in-rd/issues/126#issuecomment-739027721, or unsubscribe https://github.com/notifications/unsubscribe-auth/AE3WNSAVTRLDSFJ3SWKNDJLSTFHHRANCNFSM4UN5RK4A .