MissAssembly Correction

Question

[hafiz.talhamalik]

I am working on fungal genome assembly using long reads and I am getting lots of misassemblies. Assembly was made using NextDenovo, which made 11 contigs with lots of misassemblies. stats are as follows:

Largest contig  = 8126790
N50 = 5981938
misassemblies = 1939

I applied nextPolish as well but nothing improved. I m just wondering how can I improve this ??

[samuel.a.odonnell]

How are misassemblies identified? Have you compared the number with another assembler such as Flye recommended below? (it is fast and easy to use even just for this test)

[hafiz.talhamalik]

misassemblies were identified using quast and closest specie was used as a reference. I tried shasta, canu and NextDenovo for assemblies. Others had more misassemblies and more number of contigs, low n50 as well

[samuel.a.odonnell]

With the next closest species as a reference these 'missassemblies' could just be genuine differences.

Answer 1

[colindaven]

Which reads ? I would try preferably flye, or shasta, which are both easy to install and run

[hafiz.talhamalik]

I did try shasta but will try flye as wel.. just in case if misassemblies are still there how to overcome them ??