xCell is a novel gene signature-based method for digitally portraying the cellular heterogeneity landscape of tissues. The method uses 64 immune and stromal cell types, harmonizing 1822 pure human cell type transcriptomes from various sources. It employs a curve fitting approach for linear comparison of cell types and introduces a novel spillover compensation technique to separate them. Using in silico analyses and comparison to cytometry immunophenotyping, xCell outperforms other methods. xCell is available at http://xCell.ucsf.edu/. Tumors consist of numerous distinct non-cancerous cell types and activation states of those cell types, collectively termed the tumor microenvironment. Understanding the cellular heterogeneity in the tumor microenvironment is key for improving treatments, discovering predictive biomarkers, and developing novel therapeutic strategies. Traditional methods for dissecting cellular heterogeneity in liquid tissues are difficult to apply in solid tumors. Therefore, several methods have been published for digitally dissecting the tumor microenvironment using gene expression profiles. Recently, a multitude of studies have been published applying published and novel techniques on publicly available tumor sample resources, such as The Cancer Genome Atlas (TCGA). At least seven major issues raise concerns that in silico methods could be prone to errors and cannot reliably portray the cellular heterogeneity of the tumor microenvironment. These include reliance on data sources, narrow focus on immune cell types, cancer cells imitating other cell types, limited validation in mixed samples, biases in deconvolution approaches, difficulty in inferring closely related cell types, and reliance on reference matrix structure. Gene set enrichment analysis (GSEA) is a simple technique that can be easily applied across data types and is quickly applied for cancer studies. However, it is many times hard to differentiate between closely related cell types. In contrast, xCell integrates the advantages of gene set enrichment with deconvolution approaches. It presents a compendium of newly generated gene signatures for 64 cell types, spanning multiple adaptive and innate immunity cells, hematopoietic progenitors, epithelial cells, and extracellular matrix cells derived from thousands of expression profiles. Using in silico mixtures, xCell transforms enrichment scores to a linear scale and uses a spillover compensation technique to reduce dependencies between closely related cell types. It evaluates these adjusted scores in RNA-seq and microarray data from primary cell type samples from various independent sources. It examines their ability to digitally dissect the tumor microenvironment by in silico analyses and performs the most comprehensive comparison to date with cytometry immunophenotyping. It compares its inferences with available methods and shows that scores from xCell are more reliable for digital dissection of mixed tissues. Finally, it applies its method on TCGA tumor samples to portray a full tumor microenvironment landscape across thousands of samples. It provides these estimations to the community and hopes that this resource will allow researchers to gain a better perspective of the complex cellular heterogeneity inxCell is a novel gene signature-based method for digitally portraying the cellular heterogeneity landscape of tissues. The method uses 64 immune and stromal cell types, harmonizing 1822 pure human cell type transcriptomes from various sources. It employs a curve fitting approach for linear comparison of cell types and introduces a novel spillover compensation technique to separate them. Using in silico analyses and comparison to cytometry immunophenotyping, xCell outperforms other methods. xCell is available at http://xCell.ucsf.edu/. Tumors consist of numerous distinct non-cancerous cell types and activation states of those cell types, collectively termed the tumor microenvironment. Understanding the cellular heterogeneity in the tumor microenvironment is key for improving treatments, discovering predictive biomarkers, and developing novel therapeutic strategies. Traditional methods for dissecting cellular heterogeneity in liquid tissues are difficult to apply in solid tumors. Therefore, several methods have been published for digitally dissecting the tumor microenvironment using gene expression profiles. Recently, a multitude of studies have been published applying published and novel techniques on publicly available tumor sample resources, such as The Cancer Genome Atlas (TCGA). At least seven major issues raise concerns that in silico methods could be prone to errors and cannot reliably portray the cellular heterogeneity of the tumor microenvironment. These include reliance on data sources, narrow focus on immune cell types, cancer cells imitating other cell types, limited validation in mixed samples, biases in deconvolution approaches, difficulty in inferring closely related cell types, and reliance on reference matrix structure. Gene set enrichment analysis (GSEA) is a simple technique that can be easily applied across data types and is quickly applied for cancer studies. However, it is many times hard to differentiate between closely related cell types. In contrast, xCell integrates the advantages of gene set enrichment with deconvolution approaches. It presents a compendium of newly generated gene signatures for 64 cell types, spanning multiple adaptive and innate immunity cells, hematopoietic progenitors, epithelial cells, and extracellular matrix cells derived from thousands of expression profiles. Using in silico mixtures, xCell transforms enrichment scores to a linear scale and uses a spillover compensation technique to reduce dependencies between closely related cell types. It evaluates these adjusted scores in RNA-seq and microarray data from primary cell type samples from various independent sources. It examines their ability to digitally dissect the tumor microenvironment by in silico analyses and performs the most comprehensive comparison to date with cytometry immunophenotyping. It compares its inferences with available methods and shows that scores from xCell are more reliable for digital dissection of mixed tissues. Finally, it applies its method on TCGA tumor samples to portray a full tumor microenvironment landscape across thousands of samples. It provides these estimations to the community and hopes that this resource will allow researchers to gain a better perspective of the complex cellular heterogeneity in