The Digital Optical Chemistry System

    The Digital Optical Chemistry (DOC) System is a unique UV light projector (patent pending) that can be used to manufacture biological/chemical arrays using UV photochemistry or semiconductors using standard photoresist chemistry. In the summer of 1997, this system was created at UTSW in collaboration with Texas Instruments. The heart of the system is a device based on the Texas Instruments Digital Light Processor, a chip with hundreds of thousands to millions of mirrors under computer digital control. This system was integrated into a device that also contains a sample holder and a computer controlled reagent delivery system based on oligo synthesis technology from Affymetrix and our own MerMade system. The oligo arrays can be designed from known DNA sequence using the UTSW code, PRIMO.

    The DOC system can manufacture high density oligo arrays for DNA re-sequencing or expression studies. These arrays can be readout using a fluorescent scanner, including the UTSW MAGNA scanner or commercial units. The primary advantages over the current Affymetrix GeneChip technology is: 1) the DOC can be programmed to quickly make new unique chips with out having to make a photolithographic mask, instead it uses a ‘digital mask’, so new array sequences can be put on a chip within hours, 2) the number of array elements can reach 2,000,000 with off the shelf TI DLP hardware, 3) the machine, about the size of a Xerox machine could be replicated for use in any laboratory, 4) the arrays, currently produced on glass slides (1" x 3"), can be readout using commercial readers or the UTSW MAGNA hyperspectral imager for additional multiplexing depth.

This work is supported by the National Institutes of Health, National Cancer Institute, and the P.O'B. Montgomery Distinguished Chair in Development Biology.

Digital Optical Chemistry Team Members

• Skip Garner, Ph.D. - professor at the University of Texas Southwestern Medical Center
• Robert Balog - BME Ph.D. student at the University of Texas Southwestern Medical Center
• Kevin Luebke, Ph.D. - assistant professor at the University of Texas Southwestern Medical Center
• Gina DiMasellis, BS - biological technician at the University of Texas Southwestern Medical Center
• Amy Sek, BS - computer programmer at the University of Texas at Dallas
• David Mittelman - DCMB student at the University of Texas Southwestern Medical Center
• Alexander Pertsemlidis, Ph.D. - post-doc biophysicist at the University of Texas Southwestern Medical Center

Digital Optical Chemistry Second Generation System

    The marriage of Digital Light Processing technology (Texas Instruments) with optical deprotection photochemistry, which collectively we call the Digital Optical Chemistry (DOC) System will overcome the limitations of the traditional mask based photolithographic process and ultimately make a portable platform for the construction of unique high-density arrays. The device consists of three parts, a DLP micromirror system that selectively focuses UV light onto a glass substrate on which the reactions are done, a fluidics system that delivers the photoactivatable reagents in proper sequence and a computer with software that controls the DLP micromirror system according to the desired sequence pattern. This system could ultimately create individual spots, 20 microns or smaller in size, on glass with up to 2 million spots using DLP systems. This approach can also be applied to combinatorial chemistry problems and the manufacture of custom microelectronics. The DLP system created by TI is intended to be the next generation of high resolution, very bright, color true TV sets, computer monitors/projectors and movie projectors. For more information please visit http://www.ti.com/dlp/.

Changes for the second generation system include an automated chemistry delivery system and a suite of software applications that control DOC. Our plans are to complete the second generation system and then to begin routine production of chips. The chips will initially be used for expression studies of cancer and cardiac tissues as well as for resequencing and SNP discovery. The ability to easily make chips containing new, varying sequences will make this system ideal for rapid resequencing for diagnostics and SNP discovery, etc.

Digital Optical Chemistry Second Generation System Pictures

DOC Second Generation System. The Front of the DOC system is shown on the left.

At the top left of the DOC system is exhaust tubing that carries away fumes from the organic chemicals. To the right of the tubing is the display that shows all digital masks being sent to the DLP.

The open compartment on the left side of DOC houses all the chemicals needed to produce chips. A series of valves (not shown) are mounted on the back of this compartment to regulate the flow of the chemicals.

Directly to the right of the open compartment is a large enclosed box. The box shields the light source, the optics, the DLP chip, and the reaction chamber from any outside light. The DLP and reaction chamber are stored seperately from the light source to prevent stray UV light from affecting the chip.

Below the main DOC hardware is a waste container, various power control units, and the control computer.

The DOC Back Internal Chamber (some optics removed). The 200 watt lamp is shown at the right. Light passes through a heat absorbing filter and the short pass filter before passing through a hole in the left panel of the box and ultimately hitting the TI Digital Mirror Devic(shown in the metal case at the far right. Additional optics in the other internal chamber direct the image of the DMD to the slide holder, where oligo synthesis takes place.  This new arrangment is designed to have 5 micron optics to enable utilization of the full density of micromirrors on the DMD.

The DOC Digital Masks. The set of the masks for  6,217 oligonucleotide probes that are complementary to sequences within each Yeast open reading frame. These sequences vary in length, between 17 and 23 bases. Yeast ORF 1 (Stanford database designation) is in the upper left corner, proceeding to the right.  There are a total of 4 digital masks for each base addition, one digital mask for each C,G,A,T.

Following are the specifications and characteristics for our Digital Optical Chemistry system:

Control computer - Pentium 166 MMX PC with 14' VGA Monitor

Software - custom system software (written in Visual Basic 6 Professional)

Digital Light Processor - TI model with 800 x 600 resolution

Number of pixels - 800 x 600 = 480,000

Mirror material - Aluminum

Mirror size - 16 microns x 16 microns in a 20 micron x 20 micron spacing

Synthesis spot size - 1:1 (or less) with mirror size

Mirror switching speed - 2 ms

Reaction chamber - custom from teflon, delrin and aluminum

Reagent delivery - Liquid Dispensing System with syringe loading capabilities

Sample configuration - coated microscope slides

Microscope slide transparency - 5%@ 280, 40%@300, 75%@320, 87%@340, 88%@360, 89%@400, measured using spectrophotometer

 

DOC system diagram. The first demonstration prototype did not include the valves, so the chemicals were manually injected from syringes. The system now incorporates a valve system using a National Instruments digital I/O board and Lee microvalves.

High resolution image taken directly on Sunscreen print film. Image is 0.5" high x 0.38" wide. This demonstrates that UV light can be projected in a pattern using the micromirror system connected to a computer screen (image was produced using PowerPoint). The resolution is set by the grainy Sunscreen print film. A 365 nm filter was placed after the UV lamp to select only that wavelength, this demonstrates that the micromirror system is UV capable, as expected.

High resolution fluorescent image of a microscope slide taken in December, 1997 using a General Scanning 2000 slide scanner. This shows oligonulceotides patterened using UV projection from the DOC system. The oligos contained 2 bases (2 couplings), with the second base being marked with Cy5 dye. For scale, the black lines are 192 microns wide.

A recent chip that was custom made to re-sequence the cancer gene, p53.

Readout and associated technologies

MerMade ­

Oligo synthesis on a large scale can be done at UTSW on instrumentation developed by Dr. Garner’s group. The system, called MerMade, can produce at least 192 oligos per day at a cost of approximately $0.10 per base at a 16 nM synthesis scale. The system is designed to perform conventional phosphoamidite chemistry in an inert environment within a custom prepared 96 well plate. Reagents are dispensed from a pressurized liquid dispensing manifold from above, and are vacuum aspirated through small ports at the bottom of the wells. Each well contains a glass frit to cover the port to which CPG beads have been deposited and synthesis is performed. This custom built automatic chemistry system is computer controlled using a Macintosh computer and Labview software that are networked so that files containing oligo sequences to be synthesized are downloaded directly to the system. The MerMade system has several major subsystems - Argon pressurization system, Argon atmosphere control, reagent dispensing system, and XY motion system. Each of these subsystems is under computer control, and in particular the Argon pressurization and reagent dispensing subsystems are a component of the DOC synthesis unit.

MerMade Oligo Synthesizer. The entire system, including the control computer, amplifiers and case. On the right is a close-up of the Argon pressurized phosphoamidite bottles, the microvalve array and the injector (between the valves). The valves are mounted to the top of a plexiglass chamber filled with Argon.

PRIMO -

To complement the MerMade systems are a series of informatics tools to accurately design primers for PCR and for sequencing dye terminator reactions. The heart of this system is a new code, PRIMO, which does primer design based on a quality assessment of the template sequence from which the primers are being designed. This code avoids regions of questionable sequence quality by analyzing automated sequence output ‘trace files’ directly or by inputting quality data from the codes, phred and phrap (assembly code produced by Phil Green, University of Washington, Seattle) used in sequence assembly, thus making oligos with a higher probability of functioning well after manufacture. This code has been available for download on our www site to run both on Macintosh and on UNIX.

Some References on Chips, etc.

1. G. H. McGall, A. D. Barone, M. Diggelmann, S. P. A. Fodor, E. Gentalen, N. Ngo, The efficiency of light-directed synthesis of DNA arrays on glass substrates, Jour. Amer. Chem. Soc., Vol. 119, No. 22, June 1997, 5081-5090

A. C. Pease, D. Solas, E. J. Sullivan, M. T. Cronin, C. P. Holmes, S. P. A. Fodor, Light-generated oligonucleotide arrays for rapid DNA sequence analysis, PNAS, Vol. 91, May 1994, 5022-5026

2. G. McGall, J. Labadie, P. Brock, G. Wallraff, T. Nguyen, W. Hinsberg, Light-directed synthesis of high-density oligonucleotide arrays using semiconductor photoresists, PNAS, Vol. 93, Nov. 1996, 13555-13560

3. E. Maier, S. Meier-Ewert, D. Bancroft, H. Lehrach, Automating array technologies for gene expression profiling, Drug Discovery Today, Vol. 2, No. 8, Aug. 1997, 315-324

4. Editorial, To affinity ... and beyond!, Nature Genetics, Vol. 14, Dec. 1996, 367-370

5. L.J. Hornbeck, Digital Light Processing for high-brightness, high-resolution applications, Texas Instruments White paper, http://www.ti.com/dlp/docs/business/resources/white/index.html, and many other articles on DLP.

6. S. Rayner, S. Brignac, R. Bumeister, Y. Belodludtsev, T. Ward, O. Grant, K. O’Brien, G.A. Evans and H.R. Garner, "MerMade: A 2 x 96-Well Plate Oligo Synthesizer For High Throughput Production", accepted by Genome Research.

7. K. M. O’Brien, J. Wren, V. K. Dave, D. Bai, R. D. Anderson, S. Rayner, G. A. Evans, A. E. Dabiri, and H. R. Garner, "ASTRAL, a Hyperspectral Imaging DNA Sequencer," Review of Scientific Instruments, Vol. 69, No. 5, May, 1998.

8. P. Li, K. Kupfer, C. Davies, D. Burbee, G. A. Evans, and H. R. Garner, "PRIMO: A Primer Design Program that Applies Base Quality Statistics for Automated Large-Scale DNA Sequencing," Genomics 40, 476-485, 1997.

Related Links