[darthsequencer]

Is there a command to output the kmers of each sequence in a multifasta file?

[GenoMax]

@darthsequencer: You can use kmercountexact.sh from BBMap suite.

[roselaw27]

Trouble parsing header

Dear BBMap team:

I tried to use filterbytile.sh to remove the reads with low quality, but I encountered an error message saying that there was a trouble parsing the header. I've read the description of the script and Brian Bushnell said that was possible when the reads were renamed (such as in SRA) and to contact him if such error happened.

I downloaded the sequencing data (SRA) from ncbi and used fastq-dump to get the fastq files. I wonder if there is a solution to this?

Thank you very much!
Rose

[GenoMax]

@roselaw27: You can use `-F` option with fastq-dump to try and recreate fastq headers in original illumina format.

[roselaw27]

Thank you for your respond. Unfortunately I tried using -F and filterbytile.sh still showed the same error message. The data were from HiSeq 2500.

Thank you again. I think I am going to try to find other ways to solve the problem.

Rose

[silask]

BBsketch alltoall is incomplete

Can I ask a question about bbsketch?

I want to compare the ANI between many genomes (1000+) to each other.
I did

bbsketch.sh perfile genome_folder/*.fasta out=sketch.gz k=31,24 threads=16

comparesketch.sh alltoall sketch.gz k=31,24 prealloc=0.75 format=3 threads=16 out=table.tsv

The log files seem correct:

Set threads to 16
Loading sketches.
Loaded 1157 sketches in 59.541 seconds.
Total Time:     59.784 seconds.

Set threads to 16
Loading sketches.
Executing kmer.KmerTableSet [ways=31, tabletype=10, prealloc=0.75]

Initial size set to 45218398
Initial:
Ways=31, initialSize=45218398, prefilter=f, prealloc=0.75
Memory: max=91268m, total=91268m, free=90848m, used=420m

3.713 seconds.
Indexed 2880884 unique and 10513099 total hashcodes.
Loaded 1157 sketches in 8.457 seconds.

Ran 1225005 comparisons in 9.344 seconds.
Total Time:     17.801 seconds.

In the final output, I'm missing some genome comparisons (which I get with mash). If I run bbsketch on a subset I get the expected comparisons.


- Genomes are highly similar.

#Query Ref ANI QSize RefSize QBases RBases QTaxID RTaxID KID WKID SSU
genome1.fasta genome2.fasta 94.223 1984118 1796930 1987598 1797650 -1 -1 24.952 27.523 .

- It is not simply due to the naming: I neither find "genome1 vs genome2" nor "genome 2 vs genome1"


Any idea?

[AndrewP]

I'm trying to use BBmap to find all perfect hits or hits with an indel length 1.

bbmapskinner.sh  in=kmer.fasta out=result.sam ambiguous=all strictmaxindel=1

I'm running a control experiment where I have a subsequence to a larger sequence in the reference. I can find the subsequence in the reference if it's an exact match. However, if I add an indel in either the subsequence or reference, BBmap is unable to map the reference. I thought by setting strictmaxindel to 1, it should be able to report an alignment with a single indel. I've tried setting strictmaxindel to 10 and it still doesn't find the alignment.

Is there something that I am doing wrong?