Question: Illumina Assembly Gap Closure
4
gravatar for Lee Katz
4.7 years ago by
Lee Katz2.8k
Atlanta, GA
Lee Katz2.8k wrote:

Hi, what is everyone's experience in automated gap closure?

I have found a few papers/software that show methods of automated gap closure using paired end reads. What is everyone's experience? I've read about IMAGE, CloG, and the Reconciliator. What other software is available to try out?

assembly illumina • 4.0k views
ADD COMMENTlink modified 4.7 years ago by SES7.9k • written 4.7 years ago by Lee Katz2.8k

Does anyone have access to CloG and is willing to share? Only if the license permits though, please.

ADD REPLYlink written 4.7 years ago by Lee Katz2.8k
3
gravatar for Rayan Chikhi
4.7 years ago by
Rayan Chikhi1.2k
France, Lille, CNRS
Rayan Chikhi1.2k wrote:

In my opinion, two really competitive stand-alone gap closers are:

Your question needs to be reframed though (it appears that you want to do scaffolding). Gaps, in the assembly context, refer to sequences of undetermined nucleotides (NNN...NNN) inside scaffolds.

If you only have contigs (which do not have gaps by the previous definition) and paired-end reads, what you want is to perform scaffolding (another type of software does that). Gap-closing is a later step which consists in replacing these undetermined regions by the actual sequence.

For scaffolding, there are many options, I would recommend SSPACE or a recent one that I haven't tested, SCARPA.

ADD COMMENTlink modified 4.7 years ago • written 4.7 years ago by Rayan Chikhi1.2k

FinIS seems to require a license to a library by Mosek, so I am waiting on that.

GapCloser... are there any examples out there? It requires a scaffold file, but it seems like a catch-22. I would like to join contigs of a de novo assembly that do not have a scaffold yet.

Thank you for introducing these to me!

ADD REPLYlink written 4.7 years ago by Lee Katz2.8k

Yes that is right, you need to get a free academic license for Mosek.. Indeed, that isn't really user friendly. I just know that FinIS has a really nice algorithm.

Concerning GapCloser, the web page (http://soap.genomics.org.cn/about.html) mention that there is a manual inside the package. I successfully used a prior version, it needs the same configuration file that is used for Soapdenovo.

Your question may need to be reframed though. See my updated answer.

ADD REPLYlink modified 4.7 years ago • written 4.7 years ago by Rayan Chikhi1.2k

So basically with these programs, it is up to the user to guess the order/orientation of each contig before performing gap closing? These programs do not use PE information and contig information alone?

ADD REPLYlink written 4.7 years ago by Lee Katz2.8k

What is generally referred to as gap-closing software is a program that takes scaffolds as input, not contigs. Using PE reads, it fills the gaps present in scaffolds. Indeed, scaffolds are produced by some scaffolding software (not by user's guess), and they contain contigs in correct order and orientation.

ADD REPLYlink modified 4.7 years ago • written 4.7 years ago by Rayan Chikhi1.2k

Thank you. I think I understand better. If I understand right, I should make scaffolds with something like SSPACE or Bambus and then fill in the blanks with something like IMAGE.

ADD REPLYlink modified 4.7 years ago • written 4.7 years ago by Lee Katz2.8k

Yes. Except that IMAGE is actually an hybrid method which does scaffolding and gap-filling at the same time. You did not mention the type of data you have. But if you have Illumina, I am unsure whether IMAGE will do a good job compared to newer methods. Thus, I'd recommend using the gapclosers mentioned in the responses.

ADD REPLYlink modified 4.7 years ago • written 4.7 years ago by Rayan Chikhi1.2k
2
gravatar for SES
4.7 years ago by
SES7.9k
Vancouver, BC
SES7.9k wrote:

I'll add GapFiller to the list, which is directly compared with GapCloser in the linked paper (and is quite easy to use).

ADD COMMENTlink written 4.7 years ago by SES7.9k
Please log in to add an answer.

Help
Access

Use of this site constitutes acceptance of our User Agreement and Privacy Policy.
Powered by Biostar version 2.3.0
Traffic: 1424 users visited in the last hour