Denovo assembly combine Ilumina PE reads and sanger reads

[bobo xin]

Now,I have a denovo plasmid about 120kb，I have two types data,Ilumina 2*150 pair-end reads and assembled sanger reads(about 100K,three contigs). How can I do assembly combine these data,I see velvet -long option,but I don't know how to used it，anyone can help me???


Thanks

[sklages]

You might want to have a look at MIRA4 for this kind of project:

http://sourceforge.net/p/mira-assembler/wiki/Home/

Using velvet with '-long' is documented in http://www.ebi.ac.uk/~zerbino/velvet/Manual.pdf

[bobo xin]

You might want to have a look at MIRA4 for this kind of project:

http://sourceforge.net/p/mira-assembler/wiki/Home/

Using velvet with '-long' is documented in http://www.ebi.ac.uk/~zerbino/velvet/Manual.pdf

Thank you fou you help ,but I still have some quetion,.I used mira befor,it runs too slowly,and result is not better than velvet.

And '-long' options :
"3.2.2 Adding long reads
Reminder: you must have flagged your long reads as such when running velveth"

what does "flagged[/COLOR] your long read" mean??

[sklages]

Thank you fou you help ,but I still have some quetion,.I used mira befor,it runs too slowly,and result is not better than velvet.

And '-long' options :
"3.2.2 Adding long reads
Reminder: you must have flagged your long reads as such when running velveth"

what does "flagged[/COLOR] your long read" mean??

Have a look at the manual at chapter "3.1 Running velveth". Flagging the reads means, "make velveth know about your reads to be long reads", e.g.

./velveth out_dir 21 -fastq -long your454data.fastq

Concerning MIRA, .. assembling something small like 120KB shouldn't take too long if you take the time to have a quick look at the manual, especially about maximum coverage to use for certain sequencing technologies in assembly. There are some "DOs" and "DONTs" with MIRA. Also make sure sure that you have installed the newest version of the software (as of today 4rc4 AFAIK). Then it is a powerful assembler :-)

[ctseto]

This might be a good time to check the assembly of your Sanger reads. map your ILM reads to your sangers and see how they look in REAPr, Opera, or FRCbam. Or just visualize them in a viewer and observe if coverage is even (or if it is low), or if reads span contigs. If the mismatch rate is high between your reads and your contigs, it may suggest that combining the two may produce some complexity.

Might be worth trying to assemble with just the ILM reads to see if you can recapitulate the Sanger reads, and use that as your sanity check for the de novo assembly. Of course there will be other mis-assemblies other than your Sanger, but if a disagreement at this stage of the workflow should trigger additional scrutiny.