Meanwhile, the pH- and stress-related genes that we and others have characterized continue to appear in the literature as among the top gene expression findings in comparative studies for a variety of diseases and conditions [ 9, 23].
Similarly, in the other study [ 64], liver tissues as well as omental adipose, subcutaneous adipose, and stomach tissues were collected from 1,008 patients during RYGB (Roux-en-Y gastric bypass); 651 liver samples were gene expression profiled and genotyped, and 7,593 genes with top gene expression variance were selected for Bayesian network reconstruction.
The GENE76 model is defined as a hierarchical model using two linear combinations of the top gene expressions with respect to a ranking based on Cox's proportional hazards model.
Since there have been at least 20 genes implicated in sulfur assimilation [5], we chose to average the top 30 gene expression differences to evaluate our model.
We first independently clustered the top 300 gene expression profiles identified by the P, A, C, S and L methods (Figure 3; Supplementary Information, Table S8, S9, S10, S11 and S12 and Materials and Methods).
We needed to take the top 200 gene expression pairs by Pearson correlation from the flattened pathway to get a set of unique genes of comparable size (N = 119) to the set produced by Paradigm.
Nine of the top 25 gene expressions (IGHM, MS4A1, FCRL3, IGHG1, PAX5, BLK, IGHA1, IGL@, FCRLA; Table 2) were present in the top 100 probe sets for B-lymphocytes (see Additional file 6).
Furthermore, the TFF3 gene is among the top genes whose expression correlates strongly with that of the oestrogen receptor in large cohorts of breast cancer patients (Gruvberger et al. 2001, West et al. 2001).
(c) Top panel: Relative gene expression levels of ZEB1 and ESRP1 in epithelial (n = 11), hybrid E/M (n = 11) and mesenchymal (n = 37) cell lines from NCI-60.
Top variance in gene expression was associated with the specific metabolic requirements of the corresponding growth phase (Additional file 5).
Notably, an integration model from different sources of biological knowledge achieved 78.82% balanced accuracy and outperformed the top models with gene expression or single knowledge-based data types alone.
