RNA-seq reads were mapped using Bowtie 0.12.7 allowing up to 2 mismatches, to the library of human transcriptome sequences obtained from ENSEMBL (GRCh37.67) reference chromosomes, then entries with identical gene symbols were merged. The transcriptome includes both protein-coding genes and non-coding genes such as lincRNAs. EMSAR was used to quantify the expression levels in TPM (transcripts per million) and to infer read counts for individual genes. Differentially expressed genes were identified using edgeR 3.4.2 and confirmed using DESeq 1.8.3.
