Integration of Multiple Genomic Data Sources in a Bayesian Cox Model for Variable Selection and Prediction

Table 3

Glioblastoma data: overview of the genes among variables with the largest marginal posterior selection probabilities in the model with either the uninformative or the informative prior.

Symbol	Full name
ACSMD	Aminocarboxymuconate semialdehyde decarboxylase	0.337	0.083
AKR1B10	Aldo-keto reductase family 1 member B10		0.104
CALB2	Calbindin 2	0.080
CDH10	Cadherin 10	0.081
CNN1	Calponin 1	0.085
CRHBP	Corticotropin releasing hormone binding protein	0.234
CYB5R2	Cytochrome b5 reductase 2	0.098
C14orf132	Chromosome 14 open reading frame 132	0.154
ENPP5	Ectonucleotide pyrophosphatase/phosphodiesterase 5	0.131
FLRT2	Fibronectin leucine rich transmembrane protein 2		0.157
METTL7B	Methyltransferase like 7B		0.060
PXDNL	Peroxidasin like	0.100	0.140
RCAN2	Regulator of calcineurin 2		0.109
SLC4A4	Solute carrier family 4 member 4		0.059
SP8	Sp8 transcription factor	0.083	0.194
UGT2B7	UDP glucuronosyltransferase family 2 member B7		0.076

denotes the posterior selection probability in the model with the uninformative selection prior and denotes the posterior selection probability in the model with the informative selection prior.