 Visualizing single cell data in a T-SNEAM batting is fun. But what we really want to know is which clusters represent which cell types. In this video I will show you how to use marker genes to achieve this goal. Consider a data sample from 10x genomics on bone marrow mononuclear cells. The data on 1000 cells is available from single cell data sets widget and shows a nice clustering structure when viewed in a T-SNEAM batting. While this data comes from a healthy donor and a donor with leukemia, we will disregard this distinction here and will focus on the exploration of the cell types. We can distinguish between cell types using the expression of marker genes. For our data, genes that are expressed in monoclonal antibodies are good candidates for markers. We can find them, for example, in the cluster of differentiation marker handbook. From the book, I will use a few markers for T cells, B cells and erythrocytes. Let me put them down in an Excel spreadsheet. I will create a table with two columns, one for genes and the other one for cell types. CD3e is a marker for T cells. And so is CD8. And CD19 for B cells. From the handbook, I will use a CD235a as a marker for erythrocytes. Here is my short list of marker genes with associated cell types. Let's save it to the desktop. Let us use the file widget in orange to load the marker data. Orange correctly assumes that the cell type is a category and that the gene column contains text. I will use the genes widget to convert these texts to gene IDs and tell it that the genes are stored in the data column called genes. All our four genes were found in the NCBI database. I will display my list of genes in the data table to also show the cell type. In the data table, I can select any row and the widget will communicate the selection further down the workflow. Let me select the two rows with the T cell markers. Now I would like to score the cells in my data according to the expression of the selected gene markers. I will use the score cells widget, give it the data from single cell data sets and the list of genes from the data table. Score cells adds a column with a cell score to the single cell data. Try changing the selection of the markers to see how this alters the score. Now, we will use another T-SNE widget, fill in the data from score cells and set the color and the size of the dots to reflect the score. It looks like the cells in the lower right are the T cells. The cool part, however, is to have both the list of marker genes in the data table and the T-SNE visualization open side by side. Now, I can select any marker and see the changes in the T-SNE. The B cells are in the two central clusters and the rithrocytes are in the cluster on the left. And again, the T cells are in their own clusters. I could now add some more genes to my Excel spreadsheet, reload the data with the file widget and play around with some more markers. Or, alternatively, I could use marker genes widget to load markers from cell marker database. Or, select a cluster of cells to find differentially expressed genes and with those potential new markers. We will do all of these and much more in our next videos.