- Although therapeutics that modulate the immune system provide remarkable benefit for many cancer patients, predicting who will respond remains an unsolved problem. As interactions between the immune system and cancer are governed by a complex network of cell-cell interactions, knowing the specific immune cell composition of a solid tumour may be essential in predicting response to immunotherapy. Here, we describe how to derive the cellular composition of a solid tumour from bulk gene expression data by mathematical deconvolution, using consensus cell type-specific gene expression profiles from recently published tumour-derived single-cell RNA sequencing data. Notably, successful deconvolution depends on these new data, as previously-available profiles from peripheral blood are insufficient. The presented method makes the problem of obtaining a patient’s tumour immune cell composition from existing databases like The Cancer Genome Atlas as well as in the clinical setting computationallyAlthough therapeutics that modulate the immune system provide remarkable benefit for many cancer patients, predicting who will respond remains an unsolved problem. As interactions between the immune system and cancer are governed by a complex network of cell-cell interactions, knowing the specific immune cell composition of a solid tumour may be essential in predicting response to immunotherapy. Here, we describe how to derive the cellular composition of a solid tumour from bulk gene expression data by mathematical deconvolution, using consensus cell type-specific gene expression profiles from recently published tumour-derived single-cell RNA sequencing data. Notably, successful deconvolution depends on these new data, as previously-available profiles from peripheral blood are insufficient. The presented method makes the problem of obtaining a patient’s tumour immune cell composition from existing databases like The Cancer Genome Atlas as well as in the clinical setting computationally tractable.…
