Does anyone have a protocol or can anyone direct me to a protocol for a rolling circle (non-pcr based) approach to whole genome amplification. Thanks.

For starters I would just grab one of the manuals from one of the manufacturers (e.g.Qiagen/Repli-g-kit; GE/Templiphi, etc.). Alternatively you could just check the papers, depending on what you want to do.

There are several non-PCR based whole genome amplification methods typically referred to as "Next Generation Sequencing" (NGS), as opposed to Sanger sequencing. Some of the most common NGS systems are 454 pyrosequencing, Illumina sequencing, and more recently the Ion Torrent sequencing technology. These methods take advantages of generating large numbers of short reads around 100-200 base pairs in length, and then using high throughput computing to link together these short fragments into a single "consensus" genome.

 

Hope that helped.

There are several non-PCR based whole genome amplification methods typically referred to as "Next Generation Sequencing" (NGS), as opposed to Sanger sequencing. Some of the most common NGS systems are 454 pyrosequencing, Illumina sequencing, and more recently the Ion Torrent sequencing technology.

 

All of these require a PCR step in them. 454 uses emulsion PCR, as does the Ion Torrent and Illumina uses bridged amplification. The newer PacBio machine is the only one to my knowledge that does not.

 

These methods take advantages of generating large numbers of short reads around 100-200 base pairs in length

 

Illumina is the only strictly short read technology in there. 454 typically produces fragments ~500bp in length. Ion Torrent is dependent on the chip set you are using. PacBio produces reads in the 2-3kb range, but as up to 15% error in base calling.

 

and then using high throughput computing to link together these short fragments into a single "consensus" genome.

 

Unless you're working on viruses/prokaryotes, it's highly unlikely that you'll get adequate coverage from a single lane of any of those technologies to produce a contig. There's numerous approaches to generate data specific to answer particular questions - reduced representation libraries, RNAseq, shotgun sequencing, etc.

Indeed, I was putting the same thing together, but got caught up with something. Just to add that none of these sequencing techniques are really amplifying the DNA in any significant amount (except in some cases for library preparation, but not during the actual sequencing), hence it is not relevant to the OP.