Assembly too small for polishing?

[JWDebler]

Hi, I am trying to polish a mitochondrial assembly of just over 50.000 bp.

I keep getting this error though:

minimap2 AlKewell_mito.fasta ../nanopore/AlKewell.nanopore.all.fastq.gz > AlKewell.paf

racon -m 8 -x -6 -g -8 -w 500 ../nanopore/AlKewell.nanopore.all.fastq.gz AlKewell.paf AlKewell_mito.fasta > AlKewell_mito.racon.fasta

[racon::Polisher::initialize] loaded target sequences 0.000429 s
[racon::Polisher::initialize] loaded sequences 116.784445 s
[racon::Overlap::transmute] error: unequal lengths in target and overlap file for target AlKewell_mito!

I used the same command on the full genome assembly and that worked fine. Is my fragment too small? As it might be smaller than some of the contained reads.

[rvaser]

Hi Johannes, there are no boundaries on sequence length in Racon. Can you please paste the output of the following commands:

head -n 1 AlKewell.paf
tail -n +2 AlKewell_mito.fasta | wc

Best regards, Robert

[JWDebler]

head -n 1 AlKewell.paf
532abcbc-a198-494f-b540-cc61a136b2c5    723     15      710     -       AlKewell_mito   52512   8653    9351    504     704     60      tp:A:P  cm:i:76 s1:i:504        s2:i:0  dv:f:0.0373     rl:i:0

tail -n +2 AlKewell_mito.fasta | wc
    657     657   53826

[rvaser]

Are there carriage return (\r) characters in AlKewell_mito.fasta? Please run head -n 2 AlKewell_mito.fasta.

[JWDebler]

Wow, thanks. Yep, what a noob mistake. I exported the fasta from geneious which seems to have inserted invisible characters.

dos2unix fixed it.

Is this a problem of minimap or racon parsing the fasta? I will notify the geneious developers, as I now remember running into a similar problem in the past after exporting a fasta.

Cheers!

[rvaser]

The problem is that the parser in Racon does not remove whitespaces if the sequence is in multiple lines, I will have to fix this :)

[rvaser]

Fixed from 1.4.20 (maybe even sooner).