previous episode

Genome Bioinformatics - Computer practicals

next episode

COO-III-3. Gene content associated with niche adaptation

Overview

Teaching: 0 min
Exercises: 70 min
Questions

  • How do genomes evolve under different environmental constraints?

Objectives

  • Search for genes associated with life under different temperatures

  • Search for additional features such as CRISPR elements

Thermophiles need to survive harsh conditions that can deteriorate their DNA and proteins, or affect the activity of key biological processes. Therefore, they often acquire key genes via horizontal gene transfer.

As usually, let’s first get our files ready:

cd ~/GenomeBioinformatics/Block1/COO-III/
mkdir 06_geneContent
cd 06_geneContent
ln -s ../01_genomeData/bin.3/bin.3.faa .
ln -s ../04_genomeDensity/col_prokka/prokka_Col_02*/*faa .
ln -s ../01_genomeData/heim/prokka_GCA_0*/*faa .
rename GCA_ heim_GCA_ GCA_*
ln -s ../01_genomeData/pangui/prokka_GCA_0*/*faa .
rename GCA_ pangui_GCA_ GCA_*

Exercise: How many genes involved in repair mechanisms?

Let’s first do a general search for genes involved in repair mechanisms. What do you see?

grep -c repair *faa

Run this command without ‘-c’ to see the actual functions

Look also for other protein functions sometimes linked to adaptation to thermophily, such as repair proteins like RadA and Nre, protein stabilisation systems like the Thermosome, DNA stabilisation proteins like those involved in polyamine synthesis, or those involved in oxidative stress reduction like superoxide reductases.

Reverse gyrase

One particular protein is often a smoking gun for thermophily: reverse gyrase. While its exact mechanism is still not well understood, this protein performs a certain DNA supercoiling activity that plays a role in life at high temperatures. Here, we will try to find this protein, and evaluate its origins.

Exercise: Which genomes contain reverse gyrase?

Solution

$ grep "Reverse gyrase" *faa

All thermophilic genomes seem to contain Reverse gyrase genes. Surprisingly, also our Asgard archaeon has one copy of it.

Now, let’s extract the reverse gyrase protein sequences:

grep -h "Reverse gyrase" *faa | sed 's/ .*//' | sed 's/>//' > revGyr.list
cat *faa > all_genomes.faa
seqtk subseq all_genomes.faa revGyr.list > revGyr.faa

And reconstruct a phylogenetic tree using these, and a reference set of archaeal reverse gyrases. We will do this by using some quick tools that will give us approximate but accurate enough answers.

cd ~/GenomeBioinformatics/Block1/COO-III/
mkdir 07_reverseGyrase
cd 07_reverseGyrase
ln -s ../06_geneContent/revGyr.faa .
ln -s ~/data_bb3bcg20/Block1/COOIII/Data/revGyr_refs.faa .

cat revGyr.faa revGyr_refs.faa > revGyr_OursAndRefs.faa
famsa revGyr_OursAndRefs.faa revGyr_OursAndRefs.faa.famsa
trimal -in revGyr_OursAndRefs.faa.famsa -out revGyr_OursAndRefs.faa.famsa.tr05 -gt 0.5
perl ~/data_bb3bcg20/bin/scripts/remove_gappy_seqs.pl 0.5 revGyr_OursAndRefs.faa.famsa.tr05 > revGyr_OursAndRefs.faa.famsa.tr05.over50p

OMP_NUM_THREADS=12
export OMP_NUM_THREADS
FastTreeMP -lg revGyr_OursAndRefs.faa.famsa.tr05.over50p > revGyr_OursAndRefs.faa.famsa.tr05.over50p.fasttree

In the previous commands, you are (1) retrieving the sequences from your genomes of interest and the reference sequences, (2) placing them all in the same file, (3) aligning using the very fast aligner FAMSA, (4) removing columns with over 50% gaps, (5) removing sequences formed by over 50% gaps. Finally, you are reconstructing a quick tree using FastTreeMP. By default, this tool uses all available CPUs. To avoid that, we are modifying an environment variable from which this tool reads how many CPUs you allow it to use – in this case, 12.

Exercise: Analyse the tree

Use the command-line package newick utils to visualise the tree:

$ nw_display -w 150 revGyr_OursAndRefs.faa.famsa.tr05.over50p.fasttree | less

Use ‘q’ to quit the visualisation.

Solution

While the tree is not very robustly reconstructed, it indicates that the copies of the reverse gyrase were obtained via horizontal gene transfer from other archaea (i.e. Asgard archaea are not the sister group). It may also appear that the sequences may have originated through different events of horizontal transfer.

CRISPR arrays

One final element we can analyse is CRISPR arrays. We don’t yet completely understand why, but thermophilic tend to have a higher prevalence of CRISPR sequences in their genomes, both in terms of the fraction of genomes containing them, and in terms of the number of CRISPR repeats+spacers included in the genome. Let’s analyse the presence of CRISPR sequences in these genomes, too.

cd ~/GenomeBioinformatics/Block1/COO-III/
mkdir 08_CRISPR
cd 08_CRISPR
ln -s ../01_genomeData/*/*fna .

To do this analysis, we will use the tool MINCED, specialised in detecting CRISPR arrays:

for f in *.fna; do minced $f > $f.crispr; done

Exercise: Which genomes contain CRISPR elements?

Solution

grep Repeats *crispr

It seems that, once again, it is only thermophilic genomes, as well as our Asgard archaeal genome, which contain CRISPR elements. Looking at the number of repeats, it also appears that genomes with higher numbers of repeats are found in genomes that grow at very high temperatures.

Key Points

previous episode

next episode