Home / Healthcare & Medicine / Antimicrobial & Antibiotic Resistance / Exploring the Landscape of Antibiotic Resistance in Microbiomes / Visualization of resistome composition in ResistoXplorer
Visualization of resistome composition in ResistoXplorer
Visualization of resistome composition in ResistoXplorer
Hi, and welcome back to another video. This time, we’ll talk about how to visualise a resistome count data in Resisto Xplorer. So to start, we are here at the home page of Resisto Xplorer. You’ll click ARG Table Model. Wait for it to load. If you’ll be loading your samples, this is the form you’d use. Here, we’re going to use the pig and broiler example data set. So it’s selected already. We just click Submit.
If you want more information or details about the sample you’re looking at or the data set, you will find it here. You can read the text summary. And you can also check the library size overview.
For now, we’re going to click on Proceed to continue.
Great. Data filtration, we’ll use the default settings. So we just– it’s already selected. We just click on Submit. And here, we have the information, how the filtering went and Proceed. Normalisation now for visualisation of resistome count data, we’ll use philtre data. So we don’t have to normalise. So this, we can actually just select blank, Submit, and Proceed. So we get to our Analysis Panel home page. We’re now going to focus on compositional profiling and, more specifically, visual profiling and hierarchical analysis. So let’s start with visual profiling. Click.
Here, we have the information about the sample. So on Visual Profiling, we’re looking at the stacked bar chart. This is showing us, in each rectangle, in each colour, a different feature in each of the samples of our data set. The profile level of feature is mechanism, in this case. And the samples are grouped by species or animal host level. We can change this by clicking on the dropdown menu and selecting, for instance, country. Click on Country and Submit.
Now we see samples clustered by country instead of animal host. And you can see a different pattern here in the same country, which indicates, as you can imagine, the different animal hosts. Let’s go back to species. Click on Submit. The profile level can be changed from mechanism to class. Click on Submit. Now we see the features by class. And even further, this can be changed to gene or ARG level.
At this level, though, there can be too many bars and too many colours to look at. So you can change the threshold level from 10 to, for instance, 10,000 and click Submit. Now there are fewer bars to look at. And, potentially, interpretation gets a little bit easier.
Let’s go back to mechanism level and the threshold of 10.
Further, we can change this graph from percentage of abundance in stacked bar to stacked area plot. Click and Submit.
And we can also change from relative abundance to actual abundance.
Then, of course, data is not relative anymore.
Let’s go back to our previous case.
Further, you can organise the samples you’re looking at by groups. Let’s say species or animal hosts. Now you’ll see all of the samples together. And you can even look at individual samples if you click per sample and, let’s say, choose the third sample. Click on Submit. Now you have the composition of features for this sample specifically.
Great. So visual profiling is very helpful. One of the issues, though, is that you cannot see the different levels of profiling all in the same graph. So for instance, you cannot see mechanism class and ARG in the same visualisation area. For that, let’s go back to Analysis. So now let’s continue and click on the second item, hierarchichal. Click. So one of the advantages of hierarchical and positional profiling is that, now, we can see the different hierarchical levels all in one view. We can see mechanism class and gene level in the same view. So as default, we get the Sankey diagram. Let’s scroll down a little bit.
Here, we have the three different levels, mechanism on the left-hand side, class in the centre column, and gene or ARG level on the right-hand side. Each column or note or rectangle indicates a different feature in its class.
So one of the advantages of hierarchichal compositional profiling is that we can see all the different hierarchical levels in the same area of view. So as default, we get the Sankey diagram. Here, let’s scroll down a little bit.
Now we can see the three different levels of features. On the left-hand side, we have mechanism. In the middle column, we have class. And on the right-hand side, we have gene or ARG level. Each rectangle or bar shows a feature in its level. And the height of it indicates how abundant it is. The connection between different levels indicates how they’re related together. And the width of the connection indicates the sheer abundance between them.
We can change the way we visualise samples by, for instance, focusing on the experimental factor, looking at samples individually as well. We can change the abundance value as absolute counts or relative proportions. And we can also further change the threshold level of our samples to visualise things in a different light, similarly to what we’ve done with visual profiling before. Other forms of visualisation here include tree map. So let’s click on tree map and Submit.
Here, we have things organised at the higher level, which is mechanism. We click, and we can go further down to gene or ARG level. If we want to go back, you just click on this rectangle on top.
And the third form for visualisation is sunburst. So it’s click on sunburst and Submit.
Here, we have the different levels organised by ranks, with the innermost ring indicating mechanism, the middle ring indicating class, and the third ring indicating gene or ARG level. If we select one of them, we can single it out and analyse it in a bit more detail. If you want to go back, we just click in the centre of the diagram and go back one level to focus on another part and so on. Similarly to Sankey diagram, you can change the different factors and play around to visualise your data in different ways. This is it for the video today. We hope it’s been helpful to you and you can visualise your resistome data better now. Thank you.
In this video, we will see how we can visualise and interpret resistome data in ResistoXplorer based on the real example data set: ‘Pig and Broiler’. We recommend you follow along in your own computer and try to answer some of the questions below in the exercise section.
Hands-on Exercise:
Go to the ResistoXplorer and upload your own data or use an example dataset, such as the ‘Pig and Broiler’, in ARG table module and complete the processing step with default parameters. Once you reached the Analysis Panel page, you can see the “Visual and Hierarchical” modules under Composition profiling section. Click on each one of them and visualize the resistome composition with different approaches and answer the following questions:
- What are the differences in resistome composition between the different animal hosts?
- Do you have any biological interpretation for these differences?
- Apart from the differences between the animal hosts, what are the other striking differences between samples?
- Are the differences in the resistome composition between the ‘Pig and Broiler’ samples exists at all functional levels or only at ARG level?
This article is from the online course:
Exploring the Landscape of Antibiotic Resistance in Microbiomes
Created by
This article is from the free online
Exploring the Landscape of Antibiotic Resistance in Microbiomes
Reach your personal and professional goals
Unlock access to hundreds of expert online courses and degrees from top universities and educators to gain accredited qualifications and professional CV-building certificates.
Join over 18 million learners to launch, switch or build upon your career, all at your own pace, across a wide range of topic areas.
