Bubble plots

 

Quite frequently, you may want to provide a quick and easy to grasp plot that shows the expression of a couple of genes, segregated over a few groups. With single cell mRNA data, this also needs to take into account the fact that a gene if frequently not detected in all of the cells.

Enter the bubble plot. This visualization will show the data much like a heatmap, where every sub-group is represented by a proportional circle. The dcolor here then reflects the average intensity value, while the circle size denotes a percentage of the cells that show expression.

Within R2, we have also got this type of viaualization implemented, making it easy to use such insightful graphs with ease. How the circles are scaled can also be defined in different ways. You can look at this from the perspective of the total data set size, and represent the data as a percentage of all the cells. Alternatively, you can also represent the data as a percentage of the cells represented within a group. This later version will loose the fact that not all groups are equally big (making it harder to interpret as deconvoluted signal), but it will reflect the relative comparison between genes in the same group. 

As we encounter everywhere in science. There is not a single option that is preferred, but rather your question will coincide with the option to choose.

A step by step example on using bubble plots in R2 is also available here (R2 tutorials)

We hope this brief introduction of the bubble plot is helpful. Please give it a try in R2 and see for yourself if this is a plot you find useful.


Reacties

Een reactie posten

Populaire posts van deze blog

Travel through Transcript evolution with the Gencode version Time Machine in R2platform

Afbeelding
The definition of genes and transcripts contained within these bounderies is by no means fixed. As a scientific community we keep on learning, resolving and discovering adaptations on our gene models. The Gencodegenes consortium forms a rich resource that keeps us up to speed with the recent developments in our understanding and composition of transcripts. They do this with version controlled updates that occur a few times per year. Within the embedded Genome Browser of the R2 genomics analysis and visualization platform ( https://r2.amc.nl ), we also maintain a Gencode version Time Machine track. This track allows you to easily grasp how some transcripts are very stable, and have not changed at all, while others are still changing from gencode version to version. Within R2 we also publicly provide the RNA Atlas, which is a comprehensive collection of deeply profiled ~300 tissues, cell types and cancer cell lines, all assessed by polyA as well as total RNA sequencing. This rich resou
Meer lezen

Plot update for the R2platform

Afbeelding
  Plot update for the R2platform We have been working hard to bring you the latest update in the R2 genomics analysis and visualization platform ( https://r2.amc.nl ); the goto web-tool for biomedical researchers to test their hypotheses on public omics data, without the need for bioinformatics or coding expertise. In the latest update, we have made substantial changes to many of the plots that can be produced within R2. Many of the adaptable parameters have now become interactive and are placed under the 'gears' icon. The 'gears' icon Let's have a look at a gene within one of the many resources that are publicly available in R2. We will look at the GATA3 gene in an integrated resouce where samples from both GTEx as well as the TCGA have been integrated and 'batch corrected' to get an immersive view of expression in normal as well as cancer samples. We first select our resource of interest by clicking on 'select a data set'. and then find the 'TC
Meer lezen

Creating Raincloud plots in R2

Afbeelding
Imagine you're studying the levels of a particular gene's expression in different samples. Each cell type has its own level of gene expression, and you've collected this data for many samples. Now, you want to visualize this data in a single comprehensive overview. Let's do this using a raincloud plot. Dots or Raindrops: On the raincloud plot, you'll have a bunch of small dots, like raindrops, scattered along a horizontal axis. Each dot represents the gene expression level in a single sample. Samples with higher expression levels will have dots higher on the axis, and samples with lower expression levels will have dots lower on the axis. Cloud-like Shapes: The raindrops will be surrounded by cloud-like shapes. These shapes show you the distribution of the gene expression levels. If the cloud is wide and covers a large area along the axis, it means the gene expression levels are spread out across a wide range of values. If the cloud is narrow and tall, it indicates t
Meer lezen