Bionano assembly genome have too much N

Question

[xzpgocxx]

Hi,

I want to assemble my study genome use PacBio long sequence and Bionano with Hybrid-Scaffold way.

Firstly, I assemble the contigs use PacBio long sequence and the resulted have only 0.1% N.

Then, I assemble the scaffolds use Bionano data (Direct Label and Stain (DLS) chemistry ) as the suggestion in Bionano Solve Theory of Operation: Hybrid Scaffold. But the assembly results have too much N (nearly 27%) and the genome size is increased compared to the k-mer assess.

So, whether I ignore some parameters and what should I do for getting the best assembly results?

Any help is much appreciated. Thanks.

[GenoMax]

As I recall there are very few things one can select when doing Bionano assemblies (only things being potential genome size and if you expect a lot of repeats). Your best bet may be to talk with Bionano support (who can look at your data remotely, if you own the machine) to see if they can suggest ways of improving the assembly. Having more N's in bionano assembly is not surprising since it is trying to find a solution for fit the pieces of the puzzle (fragments seen) together.

[xzpgocxx]

Actually, I have an email to Bionano support, but the reply is confusing.

It might not be clear why i referred you to the xml as this contains descriptions that might help you to improve as all parameters are described there. How ever bionano has put a lot of effort in getting the parameters optimal and trying to improve might introduce additional errors which may look like it has gotten better but infact makes it less optimal. Running the script with the -help parameter. gives you al the options available and the other tweaking can be done in the hybridScaffold_DLE1_config.xml hopem this clarifies as well.

Kind regegards,

Kees-Jan