Gene expression (FPKM) across 91 tissues and cell types from 723 RNA-seq datasets in cattle. We computed the t-statistic to measure the tissue-specific expression pattern for each gene in each tissue and cell type. Our cattle-GeneAtlas V1 provides a primary source for GWAS interpretation, functional validation, studies of adaptive evolution, domestication and genomic improvement in livestock. For intance, you could conducted a tissue-enrichment analysis (like GO or KEGG enrichment analysis) using Cattle-GeneAtlas to detect the potential tissues or cell types on which genes of your interest (e.g., from a certain GWAS or evolution study) would act. You also could differentially weight SNPs in genes that are highly specificially expressed in the trait-relevant tissues in the prediction models (like SSGP, multi-BLUP or BayesRC) to improve genomic prediction for that particular trait, especially in the multi-breeds or over-generation situations.
This R code can be used to explore and plot the number of expressed genes and samples clustering patterns from both PCA and t-SNE approaches
This R code were used to detect the tissue-specific genes by computing the t-statistics for each gene in a tissue compared to other tissues that not belonging to the same cetagory