To distinguish the electronic and lattice contribution to the improvement of ZT separately, we computed the ZT of NWs using the lattice thermal conductivity of bulk ZnO, κ l bulk, and that of NWs, κ l MD.
In order to test whether the TE insertions are distributed in a repulsive (rarely inserted in nearby regions) or a contagious (frequently inserted in nearby regions) way in each chromosome separately, we computed the linkage disequilibrium for each pair of sites detected by analysing each possible 2 × 2 contingency table [ 13].
Hypothesizing that the quantification of a country's wealth (i.e., GDP), along with its distribution, would correlate better than either indicator separately, we computed an indicator called the public wealth index (PWI).
We use routines in the GenePattern software package [ 20, 26] and separately we compute a p-value using the Fasano & Franceschini statistical test [ 27].
In the combined HGDP-CEPH, Native American, Latino, Jewish, Asian Indian, CGP, Pacific Islander, and African data set of 5916 individuals (combined human data set henceforth), separately for each locus, we computed the fraction of individuals whose genotypes were missing (lm).
For all DR gene duplicate pairs that were detected as evolving asymmetrically by our Domain Specific Analysis (see Results), separately at each developmental stage, we computed the ratio of intersection and union of expression domains for the two paralogs.
For each subject, we computed the correlation separately for the x and y directions and for each condition and found no correlation in most cases (p-value Bonferroni-corrected at 0.0015 for 32 total conditions for each subject).
To study how the information spreads in groups of different types, we consider the values for each of the group types separately and we compute the average values at a fixed group size, to account for any dimensionality effects.
For each SNP, we separately computed the non-centrality χ2 parameter (NCP) in the first and second phase as a function of risk allele frequency, case-control sample size, and assumed effect size (genotype relative risk, GRR) (Text S1).
Moreover, we separately computed the above measures by considering either all runs, or only runs where both tools ended computations.
In order to investigate the relative contribution of the SMA and the PHC to up- vs. down-regulation of the differential feedback signal, we separately computed the percentage of signal change in the SMA and the PHC of the last 5 training runs (these runs showed the best control over the feedback signal).
