mtDNA Cyclic Sequence Alignment CSAtool

Question

[marco.barr]

Hi everyone,

I encountered an issue while running CSAtool. This is my command and report:

What’s free(): invalid next size (fast)? Maybe it’s due to the way I formatted the sequences?

I hope you can help me.

Thanks!

[Ram]

Please do not paste screenshots of plain text content, it is counterproductive. You can copy paste the content directly here (using the code formatting option shown below), or use a GitHub Gist if the content volume exceeds allowed length here.

Answer 1

[Ram]

It looks like you're running some sort of heavy analysis from an HPC login node. Login nodes are not meant for resource (RAM/CPU/time) intensive work and your sysadmin might have set limits on how much resource can be consumed in a single login. Use compute nodes to run your analysis.

[marco.barr]

I started the same analysis in parallel in sbatch slurm by setting 1 compute node, 64 RAM and 10 CPU threads but I get the same error. How many nodes should I use since I'm at the maximum usable limit for RAM and CPU for the cluster?

[Ram]

Is this the tool you're using: https://github.com/fjdf/CSA ?

[marco.barr]

Yes this is it. There is nothing written about this on the manual page

[GenoMax]

Says on the page to run as

./CSA R <multi-fasta-file>

You seem to be missing the R? Plus is the Multi_fasta.txt in root directory as you have specified (./CSA /Multi_fasta.txt)?

[Ram]

The dir location is poorly redacted. See the next line for actual path.

[marco.barr]

I simplified the path in the screenshot because it was too long but I put all the correct path... Furthermore, when I run ./CSA -h it suggests me not to put R after CSA to obtain both rotation, alignment and plot. Maybe I try to put R?

[GenoMax]

Have you checked the example files provided https://github.com/fjdf/CSA/blob/master/website/Examples.zip?raw=true and made sure your program runs properly? I just checked and had no issue running the tool with examples (which seem to be simple multi-line fasta files).

You are right in that CSA can be run in many ways depending on what you are trying to do.

$ ./CSA
[ Multiple Circular Sequence Aligner v1.11 ]
> USAGE:
        [ Rotate+Align+Image ]  ./CSA <multi-fasta-file>
        [   Rotation only    ]  ./CSA R <multi-fasta-file>
        [   Alignment only   ]  ./CSA A <multi-fasta-file>
        [Alignment Image only]  ./CSA I <multi-fasta-file>
> TOOLS:
        [  Clean FASTA file  ]  ./CSA C <multi-fasta-file>
        [Get Alignment Score ]  ./CSA S <multi-fasta-file>
        [Convert FASTA to MSF]  ./CSA M <multi-fasta-file>

You can also make sure your fasta is clean by using

./CSA C Multi_fasta.txt

[marco.barr]

I had already tried with the examples to initially familiarize myself with the tool and everything worked. Now I'll try to clean my fasta and give you updates

[marco.barr]

Hi Genomax, I'm trying to clean the fasta sequence and the output is:

[ Multiple Circular Sequence Aligner v1.11 ] Loading sequences from file </home/Multi_fasta.txt> ... (179338 bytes)  ERROR: Can't write output file

Subsequently I tried to run Alignment Image only and results:

barresi.m@login01:~/Nanoporei/CSA$ ./CSA I /home/Multi_fasta.txt [ Multiple Circular Sequence Aligner v1.11 ] > ERROR: Consensus sizes don't match

what could I try to do? Is reformatting the file so that the size of the reference matches my reads? Thanks for your help.

I also wrote to the developers of the tool