How To Merge Contigs From Two Different Assembler :

Question

[HG]

HI everyone, I am new in genome assembly. I did my Denovo assembly with Spades( SPAdes: A New Genome Assembly Algorithm and Its Applications to Single-Cell Sequencing) and i also did the assembly with GapFiller (GapFiller: a de novo assembly approach to fill the gap within paired reads) .

Now if i merge the contig file from both the program which program i can use for this job??

After merging the contig the quality of draft genome will increase or not ??

If any one have experience please help me out ..

Thank you advance

Answer 1

[Hranjeev]

We have tried GAM-NGS and obtained good results. But I'm afraid your method is not appropriate because gapfiller is used after scaffolding step, if I'm not wrong. You should try to merge contigs from abyss, velvet etc. with your SPADES contig results.

Reference: GAM-NGS: genomic assemblies merger for next generation sequencing

Other tools I know off using similar species references:

• MAIA : MAIA: Integrating Genome Assemblies (Matlab code)

• e-RGA : e-RGA: enhanced Reference Guided Assembly of Complex Genomes (Perl code)

[HG]

I appreciate your opinion . Current i have two set of contig file one from spades and another from Masurca , now i just want to check the change of N50 value if i can merge these two set. Any suggestion ???

[Hranjeev]

I believe the N50 can be improved with GAM-NGS. Please try and let us know.

[HG]

thanks .......let me check once ...then i will let you know

Answer 2

[HG]

As you suggested i am using minimus2 like this way :

$grep -c "^>" S1.seq S2.seq
   S1.seq:917
   S2.seq:1668

 $ cat S1.seq S2.seq > S1-S2.seq
 $ toAmos -s S1-S2.seq -o S1-S2.afg
 $minimus2 S1-S2 -D REFCOUNT=917

Finally i got one file named "S1-S2.afg" could you please let me know what will be next step to get contig merge fasta file??

[Manu Prestat]

I'm not a minimus2 expert. However have you tried to simply type:

minimus2 S1-S2.afg

?

By the way, I've found a new tool thanks to excellent didactic directory OMICtools : I edit my answer.

[Israel Barrantes]

Did you figure it out? minimus2 (from amos 3.1.0) does not need the ".afg" extension explicitely described

Having said that, I noticed that minimus2 has a configuration bug: it assumes that MUMmer executables deltafilter and show-coords are under /usr/local/bin. A simple modification of that will solve the problem.

Answer 3

[Manu Prestat]

Take a look at minimus2.

EDIT: and GARM.

[zy041225]

I tried GARM but failed with the error message:

ERROR: A MUM was found with a start coordinate greater than
       the sequence length, a serious error has occured.
       Please file a bug report
ERROR: postnuc returned non-zero

and

ERROR: mummer and/or mgaps returned non-zero

I've tried complie Mummer 3.22 & 3.23 with make CPPFLAGS="-O3 -DSIXTYFOURBITS", but the problem remained. Do you have any suggestions?