A positive number indicates that this dataset reports more immigrants than Özden et al.
Similarity search against public databases and conserved domain analysis of predicted protein sequences indicated that this dataset is a useful resource for comparative transcriptome studies.
Table S8 gives the number of individuals and SNPs in each of the datasets, and Table S9 indicates the dataset used in each analysis.
The differences seen with GABP highlight the potential variability in performance and seem to indicate that, for this dataset, directional scoring methods were less sensitive (SISSRS, mtc, wtd), corroborating the findings from our qPCR analysis.
Significant similarity to any one mouse muscle group was not observed for non-muscle human tissues (dataset H3), indicating this finding to be muscle specific.
This analysis showed a correlation (Pearson's Correlation Coefficient 0.80, R = 0.645) between the time of isolation of the individual isolates and the root-to-tip divergence, indicating a temporal signal for this dataset and justifying a molecular clock based phylogenetic analysis.
A separate manuscript describing this comparison is in preparation, but our analyses indicated that for this dataset, LIMMA was significantly more reliable than any other model.
However, when 10 categories are used the correlation increases to 0.998 indicating that in this dataset 10 categories were sufficient to calculate a reliable standardised rate.
However, with the DTM dataset, the R m 2 – – ratio for the two classes of markers was closer to one for SVDI compared to the other methods, indicating that for this dataset, the accuracy with SVDI was less affected by the LD between markers, compared to the other methods.
A high R-value for a dataset indicates that the dataset contains wide overlapping areas among its categories, and classification accuracy on the dataset may become low.
The high coverage of the common SNPs in a particular dataset indicates that the dataset is not as comprehensive as the other one.
