Java problem with BBtool randomreads

[KaraJC]

When I try to run the BBTool randomreads with any input, I am presented with the following Java exception:

Set genScaffoldInfo=true
Exception in thread "main" java.lang.ArrayIndexOutOfBoundsException: -17
at dna.ChromosomeArray.set(ChromosomeArray.java:181)
at dna.ChromosomeArray.set(ChromosomeArray.java:165)
at dna.FastaToChromArrays2.makeNextChrom(FastaToChromArrays2.java:478)
at dna.FastaToChromArrays2.makeChroms(FastaToChromArrays2.java:345)
at dna.FastaToChromArrays2.main2(FastaToChromArrays2.java:154)
at align2.RandomReads3.writeRef(RandomReads3.java:554)
at align2.RandomReads3.main(RandomReads3.java:338)

I then have to terminate the process with ctrl+c.

This problem is not present in any other BBTool that I have used. Can anyone help with this or suggest an alternative tool for generating synthetic reads with the same level of control offered by randomreads?

Thanks in advance.

Kara

[GenoMax]

What kind of input/output are you using for randomreads.sh (fasta/fastq)? Can you provide the command line you are using?

[KaraJC]

That output was generated with the following:

randomreads.sh ref=ref.fasta out=reads1.fasta length=150 reads=100

If I use fastq as an output or play around with any other parameters I always get the same exception.

[GenoMax]

With BBMap (v.35.59) there is no problem executing the command you give above.

Only time randomreads.sh does not produce an output is if the input fasta is very short (one sequence of ~30 bp) but I don't see any error, the program just hangs.

What version of BBMap are you using?

[KaraJC]

I am using BBMap version 35.14. The fasta file input is 42Kb in size so that shouldn't be the problem. I am going to reinstall java and then try again to see if that makes any difference.

[Brian Bushnell]

Hi Kara,

It looks like the problem here is a strange character in your fasta file that is not a standard ASCII character. It might character 239:

´

You can try doing a search-and-replace in a text editor, but extended ASCII symbols are not well defined so you may need to look at the file manually. I've fixed this so that it will not crash when given extended ASCII in fasta files; I'll upload the fixed version (35.74) later today. However, it's not a symbol that should appear in the nucleotide sequences anyway, so there's a chance that your file is corrupt.

[KaraJC]

Hi Brian,

You are indeed right about the problem being due to an unusual character. I simplified the header to remove the non-standard characters and things are working perfectly!

Thanks very much!