This release contains all new analyses, figures, and results we conducted in response to reviewer comments. The review and our initial response can be found here: https://www.biorxiv.org/content/10.1101/2021.09.02.458673v1.full (TRiP Peer Reviews). Note that our response includes a "revision plan", and this release contains these planned analyses.
In the updated analyses, we
- Examined type of MOA predicted well
- Performed feature-level reconstruction analysis
- Performed a leave intermediate performing MOAs out analysis
- Quantified Pearson correlation as an alternative to MSE for the LSA experiment
- Produced a new figure comparing individual MOAs
- We also cleaned up a lot of redundant code
We added these changes in the following pull requests: