search

Benjamin-Lee / deep-rules

Ten Quick Tips for Deep Learning in Biology
https://benjamin-lee.github.io/deep-rules/
Other
226 stars 45 forks source link

New paper on DA for scRNA-seq #127

Open tbrittoborges opened 5 years ago

tbrittoborges commented 5 years ago

Single-cell RNA-seq denoising using a deep count autoencoder

Denoising enables discovery of subtle cellular phenotypes

After having evaluated DCA against competing methods, we tested if DCA denoising could enhance biological discovery which is impossible or more challenging to obtain without denoising. Stoeckius et al. highlight the potential for integrated and multimodal analyses to enhance the discovery of cellular phenotypes, particularly when differentiating between cell populations with subtle transcriptomic differences. (...) After denoising, the two sub-populations of NK cells become visually more clearly evident based on DCA denoised NCAM1 and FCGR3A RNA expression levels .

rasbt commented 5 years ago

This is nice! I think it would be good to add a few sentences about this in ./content/10.blackbox.md I.e., saying that DL also offers new opportunities for interpreting results.

fmaguire commented 5 years ago

Assigned to tip mentioned by @rasbt

rasbt commented 5 years ago

I think "Tip 2: Use traditional methods to establish performance baselines" seems to be the best place for that imho. Do you suggest a different tip?

fmaguire commented 5 years ago

I was just basing the project assignment for where this paper fitted best based on your comment above and @tbrittoborges WIP.

rasbt commented 5 years ago

Oh I see! Never really used this "project assignment" features and thought this was a "manually written" message :)