SPAdes: does contig with node id has/refer coverage?

[bio_informatics]

Hi Members,
I'm working on bacteria genome, illumina Hiseq paired end reads.
Used:
- trimmomatic with illumina adapter file.
- spades for assembly with careful flag, bayes hammer correction.

I get 2 fasta files: contigs and scaffolds.
Contigs and scaffolds file give output in below fashion:

scaffolds-

>NODE_452_length_82_cov_291.815_ID_903

contigs-

>NODE_488_length_109_cov_1.07407_ID_975

My question:
Does cov_1.07407 and cov_291.815 refer to coverage of the contig/scaffold?

Thank you.

[akorobeynikov]

Yes, like in Velvet, this is the average k-mer coverage of a contig/scaffold. Use the last k-mer length if you need to convert to read coverage.

[bio_informatics]

Hi,
Thanks for your reply.
I do not quite understand what you meant. I have not really well read/understood on Velvet. (Have basic knowledge that it works on de bruijn graphs and so so..)

Originally Posted by akorobeynikov

Yes, like in Velvet, this is the average k-mer coverage of a contig/scaffold. Use the last k-mer length if you need to convert to read coverage.

My assembly ran on 21, 33 and 55 k-mer (default SPAdes run).

[akorobeynikov]

Originally Posted by bio_informatics

Hi,
Thanks for your reply.
I do not quite understand what you meant. I have not really well read/understood on Velvet. (Have basic knowledge that it works on de bruijn graphs and so so..)



My assembly ran on 21, 33 and 55 k-mer (default SPAdes run).

Check this http://seqanswers.com/forums/showthread.php?t=1529 for k-mer coverage. Use k=55 for SPAdes in this case.

[bio_informatics]

Thank you for directing to that page.

Indeed the below URL far better explains what I was looking for:

http://seqanswers.com/forums/showthread.php?t=6887

Merci!