Fig. 1: Schematic of the gene prioritization framework for the integrated swine omics knowledgebase.
From: A gene prioritization method based on a swine multi-omics knowledgebase and a deep learning model
The circles represent a list of candidate genes from GWAS or any other omics analysis. The rectangles represent positive training samples and negative training samples. The dotted box represents a CNN model trained by using variation counts, expression level, QTANs/QTALs number, and WGCNA module features of the training data. The output layer of the model shows the probability that the gene is a credible candidate gene by using the “softmax” function. The candidate genes with a probability >50% were denoted as credible candidate genes and can be ranked according to their probability.
