Continuing to try to find the best kmer setting to work with SparseAssemlber after the last attempt failed due to a kmer size that was too large (k 131; which happens to be outside the max kmer size [127] for SparseAssembler), I re-ran SparseAssembler with an arbitrarily selected kmer size < 131 (picked k 111).

The job was run on our Mox HPC node.

• Slurm script: 20180423_sparse_assembler_kmer111_geoduck_slurm.sh

Results:

Output folder:

• 20180423_sparseassembler_kmer111_geoduck/

Slurm output file:

• 20180423_sparseassembler_kmer111_geoduck/slurm-164530.out

This failed with the following error message:

Error! K-mer size too large!

Well, this is disappointing. Not entirely sure why this is the case, as it’s below the max kmer setting for SparseAssembler. However, I’m not terribly surprised, as this happened previously (only using NovaSeq data) with a kmer setting of 117.

I’ve posted an issue on the kmergenie GitHub page; we’ll see what happens.