These results, although biased toward models relying on datasets close to the testing datasets (B, F), usefully complement the results of intercomparisons between models.
During peer review, we can host all of the supporting information and data (with several papers having datasets close to 100 GB in size so far) and our curators make all of it available to the peer-reviewers from our FTP servers.
We produced a reference table with 1 million simulated datasets and estimated parameter posterior distributions taking the 10 000 (1%) simulated datasets closest to the observed dataset for the local linear regression, after applying a logit transformation to parameter values.
To answer the first question, the nδ = 60 000 (1%) simulated datasets closest to the pseudo-observed dataset are selected for the logistic regression and nδ = 600 (0.01%) for the direct approach.
All of above, for the dataset close to experimental electron microscopic conditions with CTF modulation and camera effect, the LO-refinement algorithm can still work effectively to improve the map quality and the reconstruction resolution of the target module.
The triplot representation shows that the number of adults is strongly negatively correlated to depth (which is the main variable structuring the dataset, close to canonical axis 1), as well as the number of juveniles but in a weaker manner.
On model M3, CA-ML is one of the least accurate methods for the 100-gene datasets, but close to the most accurate on the datasets with 500 or 1000 genes.
Proteomics data from T. gondii [19] [21], [66] and Plasmodium spp. [10], [12], [67], [68] exhibit little overlap with each other or with the computationally-defined dataset described above, probably because none of these datasets comes close to saturation sampling of the secretome, and each was motivated by different experimental rationales and employed different protocols.
As unlabeled dataset is close to negative dataset after 4 iterations, the final training set from the last iteration was used to build our classifier to extract TSPs from protein annotation in UniProt.
The proportion of dead cells estimated for both the α-factor and elutriation datasets was close to zero, while estimates for the proportion of early cells differed between the two datasets.
As seen in the table, the mean estimate of β 1 across the 1000 simulated datasets for the full dataset is close to 0.5, as expected.
