The JDRF data provides metabolomics and clinical outcomes in juvenile diabetes. This provides a biological usecase for autoPLIER and also allows a real-world comparison.
Here are some experiments we want to consider here:
[ ] Compare autoPLIER performance to the native representation and also to PLIER -- is there a prognostic gain with autoPLIER? (this will require a literature search).
The JDRF data provides metabolomics and clinical outcomes in juvenile diabetes. This provides a biological usecase for autoPLIER and also allows a real-world comparison.
Here are some experiments we want to consider here:
Other ideas?