In the last two decades, many detailed full transcriptomic studies on complex biological samples have been published and included in large gene expression repositories. These studies primarily provide a bulk expression signal for each sample, including multiple cell-types mixed within the global signal. The cellular heterogeneity in these mixtures does not allow the activity of specific genes in specific cell types to be identified. Therefore, inferring relative cellular composition is a very powerful tool to achieve a more accurate molecular profiling of complex biological samples. In recent decades, computational techniques have been developed to solve this problem by applying deconvolution methods, designed to decompose cell mixtures into their cellular components and calculate the relative proportions of these elements. Some of them only calculate the cell proportions (supervised methods), while other deconvolution algorithms can also identify the gene signatures specific for each cell type (unsupervised methods). In these work, five deconvolution methods (CIBERSORT, FARDEEP, DECONICA, LINSEED and ABIS) were implemented and used to analyze blood and immune cells, and also cancer cells, in complex mixture samples (using three bulk expression datasets). Our study provides three analytical tools (corrplots, cell-signature plots and bar-mixture plots) that allow a thorough comparative analysis of the cell mixture data. The work indicates that CIBERSORT is a robust method optimized for the identification of immune cell-types, but not as efficient in the identification of cancer cells. We also found that LINSEED is a very powerful unsupervised method that provides precise and specific gene signatures for each of the main immune cell types tested: neutrophils and monocytes (of the myeloid lineage), B-cells, NK cells and T-cells (of the lymphoid lineage), and also for cancer cells.

There are many different cells that perform highly specialized functions in the human hematological and immune systems. Due to the relevance of their activity, in this work we investigated the cell types and subtypes that form this complex system, using single-cell RNA sequencing (scRNA-seq) to dissect and assess the markers that best define each cell population. We first developed an optimized computational workflow for analyzing large scRNA-seq datasets. We then used it to find gene markers of the different cell types present in bone marrow (BM) and peripheral blood (PB). We analyzed three different single-cell datasets to find specific cell markers using this strategy: first, we searched in the CD marker genes and then in the genes encoding membrane proteins and finally in all detected protein-coding genes. This allowed us not only to confirm known CDs that best mark some cell types (e.g., monocytes, B cells, NK cells, etc.) but also to test the ability of new genes to distinguish specific cell types. Finally, we applied a machine learning method (Random Forest) to test the accuracy of the different markers we found. As a result of all this work, we have found and propose specific and robust gene signatures to identify different types and subtypes of hematological and immune cells.

In the last two decades many detailed full transcriptomic studies on complex biological samples have been published and included in large gene expression repositories. These studies primarily provide bulk expression signal for each sample, including multiple cell-types mixed within the global signal. The cellular heterogeneity in these mixtures does not allow the activity of specific genes in specific cell types to be identified. Therefore, inferring relative cellular composition is a very powerful tool to achieve a more accurate molecular profiling of complex biological samples. In recent decades, computational techniques have been developed to solve this problem by applying cell deconvolution methods, designed to decompose cell mixtures into their cellular components and calculate the relative proportions of these elements. Some of them, only calculate the cell proportions (supervised methods), while other deconvolution algorithms can also identify the gene signatures specific for each cell type (unsupervised methods). In these work, five deconvolution methods (CIBERSORT, FARDEEP, DECONICA, LINSEED and ABIS) were implemented and used to analyze blood and immune cells, and also cancer cells, in complex mixture samples (using three bulk expression datasets). Our study provides three analytical tools (corrplots, cell-signature plots and bar-mixture plots) that allow a thorough comparative analysis of the cell mixture data. The work indicates that CIBERSORT is a robust method optimized for the identification of immune cell-types, but not as efficient in the identification of cancer cells. We also observed that LINSEED is a very powerful unsupervised method that provides precise and specific gene signatures for each of the five main immune cell-types analyzed: neutrophils and monocytes (in the myeloid lineage), B-cells, NK cells and T-cells (in the lymphoid lineage).