These sets were chosen to eliminate any dependency in each test between the two half data sets based on linkage disequilibrium.
aPercentage of correlation between intensities from random half data sets.
Fourier shell correlation (FSC) calculations from two half data sets indicate 25 Å resolution for all maps based on the 0.5 cutoffs (Table 4; Figures 4 figure supplement 1E, Figure 5 figure supplement 1E, Figure 7 figure supplement 1E).
For each reconstruction, three Fourier Shell Correlation (FSC) curves are plotted: standard FSCt (blue) between two half data sets, FSCn (noise substitution cutoff 10 Å, red) and FSCtrue (green, see Chen et al., 2013 ).
To compare alternative models, we used a cross-validation approach in which we fit model coefficients on half the data set (training set) and assessed model predictions against the other unseen half (test data set) to estimate a cross-validated R (Experimental Procedures).
An exploratory factor analysis (EFA) was performed using half the data set on each instrument, using SPSS software ver. 21, followed by confirmatory factor analyses (CFA) on the other half of the data set, using Mplus.
Clustal W alignments of the arthropodan and sarcopterygian clades for SAP30L and the sarcopterygian clade for SAP30 were created, and a distance analysis with 100 "delete-half jackknife" data set tests was performed using PHYLIP as described above.
The final resolution, obtained through FSC comparison of reconstructed half-data set volumes after applying a soft mask, was ∼5 Å (0.143 criterion) for both nucleotide states.
