hello and welcome to biostat squid in this video I will give you an overview of heat Maps which are a great visualization tool for differential gene expression analysis we will cover the basics and also how to interpret a heat map in four easy steps with a real life example so let's dive in so before we explain what a heat map is let's start with some sample data now this is gene expression data so every row is a different Gene and every column is a different sample now how can we visualize the expression of so many genes across so many samples you probably guessed it already with the heat map so heat Maps allow us to visualize patterns in gene expression data okay so let's give a more specific example we are studying gene expression differences between lung cancer tissue and healthy tissue so after rna-seq pipeline we are left with something like this to simplify things we have the expression of 12 genes and eight samples the four green samples are from healthy tissue and the four purple samples from cancer tissue and every row is a different Gene the numbers show gene expression data in particular the log 2 fold change now imagine we just color coded each tile so tiles with a positive fold change would be more red the more negative the fold change the more blue the tile would be and if a gene does not change much between conditions so its differential expression is close to zero we would just give it a white color nice so now we make the numbers disappear and we have our heat map easy right so basically instead of numbers we use colors and the color and intensity of the tiles or rectangles is used to represent changes of gene expression this way we can easily visualize which genes are mostly up regulations or down Regulators across samples but how can we see patterns with so many color tiles the answer is to Cluster the ties this involves a meaningful reordering of the rows and columns and this is how we get a clustered heat map clustered heat map looks something like this the dendrograms on the sides just indicate the results of clustering both genes and samples clustered heat maps are just heat maps that are combined with clustering methods this just means we group the samples and or the genes together based on the similarity of their gene expression pattern in this clustered heat map we see all samples were clustered together in two big groups the genes were also clustered together in two big groups but we can go further down the branches and identify three Gene sets for example okay but what about the passions we were talking about before earlier um clustering helps us identify samples that are more similar to each other based on their overall gene expression patterns for example the second group of samples over here has a general up regulation of this Gene says highlighted in pink but a general down regulation of this other Gene set highlighted in blue the blue Gene says is on the other hand up Regulators in the first group of samples so we can find many more patterns in our heat map they can involve smaller Gene sets or just a few samples it really depends on our data and what we want to visualize with it in our example we expect cancer samples to Cluster together and healthy samples to Cluster together but clustering might show us unexpected and interesting groupings nice so for the very last part of our video let's have a look at a real life example I just did a quick search in PubMed for rna-seq Publications and sure enough the first one that came up had a heat map in its first figure so this is the heat map and now we're going to go step by step to interpret it so first of all we need to check the x-axis in general every column of the heat map represents a different sample so that can be cells it can be patients and there should be some kind of label to tell you what the x-axis represents and this will also give us an idea of which samples are more similar to each other um in this case it looks like each column is a cell and they were also clustered together according to cell type next we need to check the y-axis in general here we will find the genes in cases where heat Maps display gene expression data of hundreds of genes the gene names may not be displayed but in this case it looks like they filtered out their data set and they're only showing the expression of certain subsets of genes so we actually know which genes are shown right so the next step is to check our color scheme usually the lock to full change for each gene will be shown and here it doesn't specify it it does in the actual publication of course but in this case we see that upregulated genes are in red and down regulated genes are in blue and this will help us identify with a quick glance patterns of up regulators and so generally red areas and down regulated genes generally blue areas and finally we can check if we can identify any interesting patterns I guess you can clearly see there are six up regulated Gene set clusters for each of the six cell types so if you go to the original publication you will find out that the genes are actually Gene markers that help identify the cell types so this heat map just shows that the cell type annotations and the gene expression of cell type markers match nice so I hope these General tips help you interpret your own inkbox so woohoo you made it till the end squidtastic I hope this video gave you a clear overview of heat maps and how to interpret them if you like this video please let me know and also let me know what other topics you would like to cover next have a great day and see you in the next one