sibeliaZ outputs

[runxuan]

Hi,

on the Readme page, I have the impression that the output will contain two files, a GFF file "blocks_coords.gff" containing coordinates of the blocks, and file "alignment.maf" with the actual alignment.

However, i have run sibeliaz on my data twice and I only got "blocks_coords.gff" on both times with "alignment.maf" missing. There is no error message given. Have I done anything wrong here? any parameter needs to be turned on/off?

Also I want to generate a pan-genome graph in gfa format. I suppose this can be achieved through generating "alignment.maf"?

Thanks a lot!

Best regards,

Runxuan

[iminkin]

Hi,

This sounds like a bug. What is the dataset you are trying to align? Number of genomes, their size and type.

[runxuan]

Thanks a lot for your reply IIlia. I am aligning genomes from 20 different barley genotypes. Each barley genotype genome is around 5.1G bp in size. Barley is diploid with 7 chromosomes, but there are many unanchored fragments in addition in the fasta files.

Please let me know if you need further info.

Thanks

Runxuan

[dibsondivya]

I'm currently having the same issue with Conda installed Sibeliaz 1.2.7. I'm unable to run sibeliaz due to errors but I can run twopaco and sibeliaz-lcb individually. However, sibeliaz-lcb only produces .tmp files and blocks_coords.gff in the output directory, with the alignment file empty.

[iminkin]

Hi @dibsondivya , thank you for your interest. Can you run sibeliaz on toy genomes (example1.fa and example2.fa) from this folder? https://github.com/medvedevgroup/SibeliaZ/tree/master/examples

[dibsondivya]

Hey @iminkin, I've tried running the command 'sibeliaz genome1.fa genome2.fa' but received the following error message:

/opt/homebrew/Caskroom/miniconda/base/envs/sibeliaz/bin/sibeliaz: line 107: free: command not found
stat: illegal option -- c
usage: stat [-FLnq] [-f format | -l | -r | -s | -x] [-t timefmt] [file ...]
/opt/homebrew/Caskroom/miniconda/base/envs/sibeliaz/bin/sibeliaz: line 111: total_size + : syntax error: operand expected (error token is " ")
Constructing the graph...
PARSE ERROR: Argument: (--filtermemory)
Couldn't read argument value from string '-o'

Brief USAGE:
twopaco {-f |--filtermemory } [-o ] [--test]
[--tmpdir ] [-a ] [-t ] [-r
] [-q ] [-k ] [--] [--version] [-h]
...

For complete USAGE and HELP type:
twopaco --help

Loading the graph...
error: Can't read the input file
rm: ./sibeliaz_out/de_bruijn_graph.dbg: No such file or directory
Performing global alignment..
bash: line 4: free: command not found
unlimited
find: -printf: unknown primary or operator

When I tried running twopaco first and then sibeliaz-lcb, with the commands 'twopaco genome1.fa genome2.fa -k 15 -o twopaco_test --tmpdir twopaco_tmp --filtermemory 8.6' and 'sibeliaz-lcb genome1.fa genome2.fa -o sibeliaz_test_output --graph twopaco_test -k 15 -a 150 --chunks 5'. Running sibeliaz-lcb without the --chunks command always lead to a segmentation error.

The .maf file was produced in a directory 'sibeliaz_out' but was produced empty. The contents of the .maf read:
''

##maf version=1

sibeliaz v1.2.7

cmd=genome1.fa genome2.fa

''

[iminkin]

Hi, what system are you using? Is it macOS?

[dibsondivya]

Hey! Yes, it's macOS Sonoma 14.5.

[iminkin]

It seems that SibeliaZ does not currently work under macOS. It should be an easy fix, and I will make a release soon. In the meanwhile, you can try to run it in a Linux computing environment if you have access to one.