Skip to main content

Table 5 Log(ods) and relative Pvalues associated to different single species (SSs) in logistic models

From: In silico prediction of lncRNA function using tissue specific and evolutionary conserved expression

   Univariate Bivariate Multivariate
Predictor Samples Log(odds) AUC Pvalue Log(odds) AUC Pvalue Log(odds) Pvalue
ggallus 17 –2.6 0.67 2.2e–217 –1.9 0.72 5.9e–106 –1.3 1.1e–51
ggorilla 12 –2.9 0.65 3.7e–162 –0.99 0.78 4e–19 –0.0092 0.95
hsapiens 59 –4.3 0.7 1.4e–300 –1.9 0.74 7.3e–54 –1.7 2.6e–31
mdomestica 20 –3.1 0.68 1.5e–225 –1.8 0.73 1.2e–77 –1 3.3e–23
mmulatta 14 –2.7 0.65 1e–175 –1.1 0.8 2.1e–25 0.38 0.002
mmusculus 49 –4 0.69 6.4e–275 –2.1 0.78 1.9e–74 –1.4 1.2e–24
oanatinus 19 –2.9 0.68 2.5e–229 –1.9 0.78 3.9e–92 –1.4 1.3e–46
pabelii 10 –2.8 0.64 3.7e–142 –1.2 0.81 1.7e–27 –0.53 2.2e–05
ptroglodytes 28 –3.1 0.67 1.5e–204 –1.2 0.77 1.6e–27 –0.36 0.01
xtropicalis 13 –2.7 0.65 5.5e–176 –1.8 0.78 4.4e–79 –1.5 5.3e–56
  1. Bivariate models include two predictors: the indicated species plus AS. Note that AS and SSs derive from different expression datasets, GTEx and Necsulea respectively. The bivariate model that include hsapiens (from Necsulea) and AS shows that the contribute of hsapiens to the prediction is significant even if derive from the same specie of AS. The multivariate model consider all SSs but not AS, the AUC in this case is 0.77, the same AUC that we obtain with AS alone