I want to assemble a linear genome and I have a reference genome too.

I ran the following commands for de novo velvet assembly:

velveth auto 31,45,2 -fastq -shortPaired1 -separate 2-CA-SVA-2_S2_L001_R1_001.fastq 2-CA-SVA-2_S2_L001_R2_001.fastq

I chose auto_33 for further analysis

velvetg auto_33 -exp_cov auto

Now my output files inside auto_33 folder are:


But I have no clue how to move ahead. Can I do anything else with velvet or do I have to switch to another program?

  • $\begingroup$ Blat the contigs $\endgroup$
    – SmallChess
    Commented Nov 5, 2017 at 0:26
  • 1
    $\begingroup$ Would you mind elaborating ? $\endgroup$
    – L R Joshi
    Commented Nov 5, 2017 at 0:28
  • $\begingroup$ How to proceed depends entirely on what you want the assembly for. Are you looking for variants? Is this the genome of a related species which has not been annotated and you want to find its genes? Are you looking for one gene in particular? For syntenic regions? We can't tell you how to proceed if you don't tell us what you need. Please edit your question and clarify. $\endgroup$
    – terdon
    Commented Nov 5, 2017 at 23:13

1 Answer 1


The contigs.fa file is a FASTA formatted file of the final contigs generated by velvet. I would start assessing the quality of the assembly that was generated. Here are some questions to get you started. Really the next steps depend on how you plan to use the assembly.

What is the distribution of contig lengths? Average and N50 contig length

How does the assembly compare to the reference?

  • How much of the reference does the assembly cover?

When you align the original reads back to the contigs how well are they covered?

  • Do any of the read alignments not make sense?

  • Are there sections of the contigs that are not covered?


Your Answer

By clicking “Post Your Answer”, you agree to our terms of service and acknowledge you have read our privacy policy.

Not the answer you're looking for? Browse other questions tagged or ask your own question.