With the present simulations, we have tried to create conditions that are typical for data analyses in medicine and life sciences, i.e. small to moderate sample sizes, small to moderate correlations among the variables and small to moderate amounts of missing data.
The method performed well in detecting differential network connectivity in simulations of moderate sample size, compared to other simple methods with Pearson correlations or partial correlations only.
When determining internal consistency, Wallace et al. [ 26] used a moderate sample size but presented only a Pearson's product moment correlation matrix.
Among dug well samples, moderate correlation exist between Mg and TDS, EC and Ca; TH and TDS, EC.
The pairwise sampling correlations were moderate and mostly negative (Table S1).
Porosity and P-wave velocities show moderate correlation: samples with higher porosities have slower elastic wave velocities, as seen in Figure 8C.
Similarly, tube well samples display moderate correlation between HCO3 and TDS, EC and Na; Mg and EC, K and Li; Ca and K (Table 2).
Paired pleural effusion and serum samples showed moderate correlation of soluble syndecan-1 (r = 0.45, P = 0.05).
However, comparison of the log2 fold change of the mRNA between ER+ vs. ER− sample groups that corresponds to these 479 differentially expressed genes and the log2 ratio copy number difference of the corresponding genomic coordinated in ER+ and ER− sample groups showed only moderate positive correlation (correlation coefficient = 0.28).
In both raw and cooked samples, moderate to strong correlations were observed between iAs, roxarsone, and the unknown arsenic species, whereas correlations of these arsenic species with DMA were relatively weak (coefficients ranging from 0.11 to 0.36) (Table 3).
