How to obtain synteny blocks using SibeliaZ

Overall idea is to use SibeliaZ for making alignment and then construct longer synteny blocks with maf2synteny module as described here.

0) Data Preparation

Let's suggest you already have your genomes in fasta format in merged.fna file.

We recommend installing SibeliaZ with a conda:

conda install sibeliaz -c bionconda

We recommend to use some parameters:

-k 15 k-mer size for bacterial size genomes (recommended in SibeliaZ documentation for bacterial genomes);
-a N*20 where N is the number of genomes, for detecting highly duplicated blocks (20 equal to 10 duplications within genome).
-n to skip the alignment with nucleotides and only output coordinates of the alignment saving time and memory;
-t 4 optional, for desired threads number;
Do not change -m parameter. It will slow down the computation significantly and will not give you blocks. Blocks need to be obtained with maf2synteny tool.

Final command:

sibeliaz -k 15 -n -o sibeliaz_out merged.fna

Now you have blocks_coords.gff in output folder!

We recommend using fine parameters for merging alignments into blocks.

For this, you need to create file fine.txt with this content:

30 150
100 500
500 1500

Also, create file fine_500.txt if you want to get shorter blocks (shorter then 1500):

30 150
100 500

More about these parameters and what they mean here.

Then, you can run maf2syneny SibeliaZ module with desired minimal block size -b.

For getting blocks with minimal size 5000:

maf2synteny -s fine.txt -b 5000 blocks_coords.gff

For getting blocks with minimal size 1000:

maf2synteny -s fine_500.txt -b 1000 blocks_coords.gff