Wow there is a lot of choice when it comes to buying something as simple as an NGS library prep kit. With Illumain in mind (as always) I found at least 17 providers. includding: Affymetrix, Agilent, Beckman, Bioo, Enzymatics, Epigentek, Illumina, Kapa, Lucigen, Mclab, Milipore, NEB, NuGen, Qiagen, Rubicon, Swift and Thermo. All these providers offer at least one, if not a whole range, of NGS kits for DNA; and some offer RNA, ChIP, Methlation, etc, etc, etc.
Would you have believed it Affymetrix offer an NGS library prep kit too! Their Prep2Seq kits is basically a reformulation of a standard Illumina library prep and includes the same indexed adapters (even missing out inde numbers 17 and 24).
In my lab we use Illumina and Rubicon library preparation kits for most experiments. We choose Illumina as we get end-to-end support and everything is included (now they are shipping their own beads). Rubicon allows us to access very small amounts of DNA e.g. the ctDNA exomes in Murtaza et al.
How do they work: Most of the kits use the standard end-repair, adapter ligation workflow we've all come to know and love. Many have streamlined the standard workflow and combined steps. Nearly all have moved to bead-based cleanup and consequenlty DNA inputs can be as low as a singel-cell, or 100ng without PCR.
How do they compare: I'm not aware of a comparison paper that looks in detail at kit performance. Given that library prep costs are increasingly become an expensive part of the process it's about time we did exactly that.
How much do they cost: The kits webpages do not all offer pricing information without askig for a quote, however more than half do so I used the list priocies to try and come up with a cost per sample. Some of the kits do not include adapters for library prep so you'd need to source these and include that in your costs.
Kits are between £15-55 per sample, making the most expensive kit (Millipore MagnaSeq) 3.75x more than the cheapest (NEB NEBNext Ultra).
We'll certainly be looking into kit costs in my lab as we're processing larger and larger numbers of libraries. It might even be cost-effective to go back to home-brew methods, especialy if we're after a $10 prep for a $1000 genome!