How to visualize MAGs

Question

[ellenphante]

I have recently analyzed my metagenomic data using the metagenome-atlas. I have a set of MAGs of varying completeness/contamination and from different lineages and am looking for a way to visualize the coverage.

I have previously used BRIG before but not for coverage mapping, but I am aware that it does have a coverage mapping module. However, I am confused as to what my reference data would be.

I have also had some R packages recommended to me by colleagues (ggbio, circlize) but have no experience using these.

Any suggestions on what to use? Unfortunately I only have the atlas outputs because I do not have access to a server which can operate metagenome-atlas anymore.

Answer 1

[Mensur Dlakic]

If you don't have coverage information, that can be obtained on a per-contig basis by mapping the sequencing reads onto your metagenomic assembly.

Assuming you have some kind of embedding information for your bins - say coordinates from t-SNE or UMAP embedding - it becomes simply a matter of creating a scatter plot and coloring each dot by coverage values.

[ellenphante]

Looks great - do you have any tool suggestions to do this? I currently don't have coverage stats.

[Mensur Dlakic]

There are many tools to calculate contig coverage. I will recommend calculate_read_coverage.py from Autometa.

Making a plot like I did above is fairly trivial once you have [X, Y, coverage] columns. It is a scatter plot of [X, Y] colored by coverage values, which can be done using many tools such as pandas.plot.

Answer 2

[gamamiguelangel]

If you want to analyze the coverage of you MAG trought the metagenomes you should use Anvi´o :) .... and if you want to see very quickly you can use anvi snakemate pipeline for metagenome analysis

https://merenlab.org/2018/07/09/anvio-snakemake-workflows/#metagenomics-workflow