SOAPdenovo Scaffolding

[kaboroevich]

Hello all,

I am getting some strange results for the SOAPdenovo scaffolding process. I discovered that if I use the ouput of the prepare (http://soap.genomics.org.cn/down/prepare.tgz) tool, I get different, arguably better, results.

After error correction (http://soap.genomics.org.cn/down/correction.tar.gz) I ran pregraph and contig with a kmer of 49. This generates “original.contig”. Following the normal steps, of map and scaff results in “original.scafSeq”. However, after remapping with BWA I noticed what seemed to be strong evidence for the scaffolding of many contigs.

So, I copied “original.contig” to a new directory and ran “prepare –g rescaf –K 49 –c original.contig”. Running map and scaf on rescaf.contig results in the file rescaf.scafSeq. The following is a summary of the assemblies.

File              Contigs  Total_length Longest   N50
original.contig   2459270     144845786    5254    54
rescaf.contig     2459270     144845786    5254    54
original.scafSeq    15289       4964403    5254   532
rescaf.scafSeq       9875       5416552   36403  2579

Does anyone know the explanation for this difference? I assume there must be some information generated by pregraph and contig that prevents the scaffolding.

Regards, Keith