Primer sets were designed to amplify a region within a single exon

Sample quantity and integrity was checked using RiboGreen analysis and an Agilent 2100 Bioanalyzer or Caliper equivalent. Each sample passing quality control was used to create a polyA+ stranded barcoded RNAseq library using standard protocols. Five samples were pooled for each lane for Illumina Hi-seq 2000 sequencing to generate 2040 million mapped paired-end reads per sample. Data Analysis Quality control of fastq data was performed using FastQC:Read QC. The 170 bp pairedend reads were then aligned to human genome via Tophat2 using the iGenome human UCSC reference annotation to align and quantitate the transcripts. Transcript assembly and transcript abundance was determined using Cufflinks program where the hg19_genes_2012-03-09.gtf file was used as reference annotation guide to determine fragments per kilobase per million reads for each transcript. The raw FPKM values were normalized following a previously described procedure. Principal components were computed using the surrogate PubMed ID:http://www.ncbi.nlm.nih.gov/pubmed/19778733 variable approach. We analyzed the log transformed normalized FPKM values using a linear mixed effects model adjusting for age, PF-562271 gender and the first two PCs. We tested for gene effect using the Kenward-Roger approximate F-test to account for the potential impact of small sample size. DEGs were identified by controlling false discovery rate at <0.01 accounting for 15,669 genes. These genes were in the top 75% of variation of expression and this was done to remove genes with low levels of variation. DEGs were those with a 2 or greater fold change in expression, compared to pre-transplant expression. DEGs were further investigated for function and pathway 3 / 14 Differentially Expressed Genes after Kidney Transplant enrichment using Ingenuity Pathway Analysis by analyzing genes with higher levels than baseline, lower levels, or combined higher and lower level genes at each time point. Results Transcripts expression over time after transplant Patient characteristics are described in Alignment and mapping of RNA sequencing reads Fastq files of all 96 samples passed FastQC report with good per base sequence quality, per base sequence quality score, per base N content and good sequence length distribution. RNA sequencing paired-end reads of approximately 170 bp were aligned to the reference human genome using the TopHat2 algorithm with results summarized in Functional annotation of DEGs and Ingenuity Pathway Analysis One week post-transplant. Functional annotation of DEGs with higher and lower levels with FDR < 0.01 and 2 or greater fold change compared to baseline, indicated that the top transcripts were in T-cell related pathways. This may be related to patients receiving T-cell depletion induction PubMed ID:http://www.ncbi.nlm.nih.gov/pubmed/19775575 with thymoglobulin and a calcineurin inhibitor in the immediate peritransplant period. Specifically, the top altered pathway was T-cell receptor signaling based on 18 genes with lower levels than at baseline. This pathway primarily contained the CD3D, CD3E, and CD3G genes for T-cell receptor signaling components. When all significantly higher and lower level DEGs compared to baseline were analyzed together no pathways based on higher level genes were identified due to a masking of the effect by the greater number of lower level genes compared to higher level genes. Thus, we investigated only the DEGs that had increased levels in blood compared to baseline, separately with pathway analysis and identified the axonal guidance signaling pathway containing 12 genes. Inter