The SALMON package (17) was used to estimate expression values using whole transcripts annotation from Gencode (vM22) as a reference (18). Expression level of a protein-coding gene was measured by TPM (transcripts per million) by pooling all annotated transcript isoforms of the gene with an option “-g” in SALMON. The TPM (log2) follows a bimodal distribution and a threshold of two separated actively expressed genes from those expressed at background level. A gene was defined as differentially expressed by GFold (19) if a “Gfold” value (log2) is higher than one, meaning that the probability of fold change in expression being larger than two is 0.99 with default settings. We further required DE gene to be actively expressed in at least one of the compared conditions. Heat map visualization of gene expression values were done by MeV (20).
Gene ontology enrichment analysis was done with the online DAVID bioinformatics resource 6.8 (21). Genes that were actively expressed in at least one of the analyzed cell types was included as background. Redundant GO terms output from DAVID were filtered by REVIGO (22) with default settings.
Do you have any questions about this protocol?
Post your question to gather feedback from the community. We will also invite the authors of this article to respond.