In this section, first we assessed the sensitivity of risk prediction to different LCA specifications.
If predictions belong to different domain regions, both predictions were retained.
The results focus on the range of the tsunami predictions due to different source parameters (rather than individual results).
We reported high values of Jaccard similarity index between predictions corresponding to different P-value thresholds (>0.97, see Supplementary Methods and Supplementary Table S2).
Amongst the remaining 29 DUF families, where predictions belong to different SCOP folds, some included examples such as 6-bladed β-propeller, 7-bladed β-propeller and 8-bladed β-propeller which are believed to be evolutionarily related [ 55].
In order to test the transferability of the prediction method to different MHC proteins, we trained the scoring method on binding data for DRB1*0101 and used it to make predictions for multiple MHC allotypes with distinct peptide binding specificities including representatives from the other human class II MHC loci, HLA-DP and HLA-DQ, as well as for two murine allotypes.
To develop well-adjusted prediction models to different process states, k-means clustering method is integrated with ANFIS technique.
We apply our function prediction strategy to different sets of species, ranging from species pairs to groups of up to four species.
We apply a modern state estimation algorithm (the Local Ensemble Transform Kalman Filter), previously developed for numerical weather prediction, to two different mathematical models of glioblastoma, taking into account likely errors in model parameters and measurement uncertainties in magnetic resonance imaging.
To identify which category of genes could be responsible for these differences, we used OrthoMCL predictions to investigate how the different categories previously annotated were shared among lineages.
