Compressing gene expression data using multiple latent space dimensionalities learns complementary biological representations

jaybee84 / ml-in-rd

Manuscript for perspective on machine learning in rare disease

Other

2 stars 1 forks source link

Compressing gene expression data using multiple latent space dimensionalities learns complementary biological representations #57

Closed allaway closed 4 years ago

allaway commented 4 years ago

@jaybee84 made a comment on #44 (that ended up going unaddressed) about whether we could provide some criteria or a rule of thumb for selecting dimensionality reduction methods.

This paper from @cgreene and others seems like it would be relevant to address this idea :)

Compressing gene expression data using multiple latent space dimensionalities learns complementary biological representations https://doi.org/10.1186/s13059-020-02021-3

allaway commented 4 years ago

Self assigning this to add it to the heterogeneity section.

allaway commented 4 years ago

My read is that the takeaway recommendation is that trying multiple methods is the safest bet, rather than relying on one method. It's a bit of an assumption on my part that this rule of thumb would apply to all rare disease applications as well, but I think this is a reasonable (and conservative) assumption.