Computational protocol: Mutational and putative neoantigen load predict clinical benefit of adoptive T cell therapy in melanoma

Similar protocols

Protocol publication

[…] RNAseq data were processed as previously described using TopHat2 and Cufflinks v2.1.1. Isoform FPKMs were summed up to obtain gene-level expression. Data were quantile-normalized and log-transformed by log2(data + 1). Genes were median-centered and reduced to protein-coding genes defined by the HGNC. SAM analysis was used to rank genes based on differential expression scores, DAVID was used for GO-term analysis, and GSEA with C2, C6, and C7 gene lists was used for GSEA. Single-cell signatures and melanoma lineage signatures were from Tirosh et al.. The MHC-I APM displayed high correlation of gene expression. In particular, HLA-A, HLA-B, HLA-C, TAP1, TAP2, NLRC5, PSMB9, PSMB8, and B2M were highly correlated, and further termed the “core” MHC-I set (Supplementary Fig. ). The mean expression of these core genes is the MHC-I score and was then divided in quartiles across patients to test the association with OS. For the MHC-I score quartiles, the resulting bins were: 4th quartile [−2.98,−0.493], 3rd quartile (−0.493,0.196], 2nd quartile (0.196,0.728], and 1st quartile (0.728,2.16]. The MHC-I score was subsequently applied to the TCGA and Cirenajwis et al. cohorts. To stratify the patients we again used the MHC-I score quartiles determined in the respective data sets. We applied the IPRES signatures to our cohort as specified in Hugo et al.. Briefly, 21 of 22 validated IPRES signatures were available, whereof 15 gene sets were available from Broad MSigDB ( and six gene sets were available from Supplementary Data from Hugo et al.. The gene set variation analysis (GSVA) scores were calculated from uncentered gene-expression data; GSVA scores were transformed to z-scores and the mean z-score of the signatures was obtained. A cutoff of >0.35 was applied as described in Hugo et al. for a sample to be called “IPRES-enriched”. […]

Pipeline specifications

Software tools TopHat, Cufflinks, DAVID, GSEA, GSVA
Application RNA-seq analysis
Organisms Homo sapiens
Diseases Melanoma