454 data

[nitinkumar]

what you guys thinks of a bacterial data which consist 11 scaffolds, 629 contigs with 7.5 MB (no of contigs). please tell me the criteria for choosing the correct data like what should be the size of contigs, scaffolds etc.

[flxlex]

There are no 'shoulds' here. Ideally, you would get a gapless contig per chromosome (or plasmid) but that is usually not possible. Next best is a single scaffold per chromosome/plasmid with as few gaps as possible and long as possible contigs..

How far you can get depends on the read dataset (amount, type, type of paired ends/mate pair reads).

In your case, 11 scaffolds is not bad for starters, I can't say much more than that based on the limited data you provide.

[kmcarr]

What is the average depth of coverage for your assembly? Look at the 454ContigGraph.txt file. For each contig it lists the average depth as the last column in the summary information. Ideally this should be ~20-25 for a de novo assembly. Being significantly lower than that can result in decreased contiguity. If your coverage is low you could collect additional sequence data an try a new assembly.

[Kasycas]

Was it shotgun or paired end/mate pair reads? I've found that 8kb mate pair data on my 4.5Mb bacterial genome works wonders (5 scaffolds). This was done on a 1/4 plate.

Is your organism 7.5Mb and is this close to what the scaffold covers?

[nitinkumar]

Data consist of shotgun reads and 8 Kb paired ends. the coverage of some of the contigs is lower than the 20. my reference genome is 7.5 Mb and all the scaffolds show the relationship with the chromosome and large plasmids using BlastN especially the biggest scaffold. reference genome is R.leguminosarum 3841. there are gaps in each of scaffolds, Now I want to close the gaps using bioinformatics tools. Can we use MUMmer for this job? One more question that in Newbler2.5 we are getting the N50 scaffolds = 927736, 2. what is this 2 means here??