draft genome quality

[buthercup_ch]

Hi everyone.
I just trying to deal with the work started for one of the master students in my new lab. I'm now dealing with a set of 64 bacterial draft genomes, and the first thing that I need to understand is if the quality of the assembly is good or not.
I have a huge list of different statistics, that I supposed were obtained by using Velvet for genome assembly. These statistics include the following:
- Read length
- Total read length (bp)
- Total read length (Mb)
- Estimated coverage
- Velvet hash value
- Total number of sequences
- Total number of contings
- N50
- Length lof longest contig
- Total bases in contigs
- Number of consigns > 1K
- Total bases in contigs > 1K

Can any one explain me what's the meaning of each statistic and where I should relay most to evaluate the quality of the obtained draft genome??
I'm completely lost!

Thank you very much in advance

[mastal]

The problem with the statistics is that none of them tell you what you really want to know - how accurate is the assembly.

The n50 is probably the most used statistic, but there are other statistics that give important information. Have a look at this paper,
De novo genome assembly: What every biologist should know.

http://www.nature.com/nmeth/journal/v9/n4/full/nmeth.1935.html?WT.ec_id=NMETH-201204

If the assemblies were done using velvet, you might also want to have a look
at the velvet manual:

https://www.ebi.ac.uk/~zerbino/velvet/Manual.pdf

If you need to know more about de novo assembly, have a look at this article which used to be on the SEQanswers SeqWiki

http://en.wikibooks.org/wiki/Next_Generation_Sequencing_(NGS)/De_novo_assembly