Summary

The project goal is to design, build and operate an open-source DNA synthesizer. There are many interesting DNA synthesis projects that could be pursued given the availability of large quantities of cheap large-scale highly-parallel DNA synthesis. Ultimately the chemistry of phosphoramidite synthesis will guide and inform the machine design, but so far the best direction seems to be using an inkjet for depositing 100s of millions of drops per second of chemical reagents onto a surface.

Each drop is a "pixel" that acts as a unique reaction chamber for the reagents. Each layer of the image builds upon the surface-bound reaction results of the previously deposited reagents. Different DNA molecules are formed inside each droplet pixel region on the surface. The DNA molecules are covalently bonded to either the surface or to microbeads on the surface. The droplets don't intersect or merge their contents together because of surface tension and distance.

The inkjet can target each drop and deposit different chemical reagents into every drop on every "round" of the synthesis cycle. Because of side-reactions, moisture and other errors, the length of the DNA molecule that can be manufactured inside of each location is limited to perhaps 100 to 200 bp at most, and usually must be at least 20 bp to be of any utility. Conjugation and ligation is necessary to combine short DNA molecules into longer DNA molecules. Although the synthesis cycle is well-known, the design for mechanically moving the separate strands together remains to be determined for this DNA synthesis machine. Once short DNA molecules are placed together, reactions such as Gibson assembly or Ligase Cycling Reaction or extension PCR can be used to combine short DNA molecules into longer DNA molecules. The POSAM synthesizer was designed to create DNA hybridization arrays, where the DNA would be left on the surface and not combined into longer DNA molecules. Perhaps the machine's design will use really tiny pipette tips to transport short DNA from multiple drops to a single reaction chamber?

The conjugation of these DNA molecules is critical to the formation of superlong DNA molecules between 5K bp (basepairs) and 1M base pairs. The human genome has approximately 3B base pairs, and DNA synthesizers from the 80s were making 100 bp molecules at best.

Note that DNA synthesis and DNA sequencing are two distinct procedures. Sequencing is reading, synthesis is writing. The cost of DNA sequencing (reading) has been falling quickly over the past decade, but synthesis has not moved as much.

The project is self-funded and communication with the team can be achieved by looking at the hplusroadmap IRC channel.